foreign exchange markets chapter and exchange rates markets/materiale... · 14.2 functions of the...

Salvatore c14.tex V2 - 10/18/2012 1:15 P.M. Page 423

Foreign Exchange Marketsand Exchange Rates

chapter

LEARNING GOALS:After reading this chapter, you should be able to:

• Understand the meaning and functions of the foreignexchange market

• Know what the spot, forward, cross, and effectiveexchange rates are

• Understand the meaning of foreign exchange risks,hedging, speculation, and interest arbitrage

14.1 IntroductionThe foreign exchange market is the market in which individuals, firms, and banksbuy and sell foreign currencies or foreign exchange. The foreign exchange marketfor any currency—say, the U.S. dollar—is comprised of all the locations (such asLondon, Paris, Zurich, Frankfurt, Singapore, Hong Kong, Tokyo, and New York)where dollars are bought and sold for other currencies. These different monetarycenters are connected electronically and are in constant contact with one another,thus forming a single international foreign exchange market.

Section 14.2 examines the functions of foreign exchange markets. Section14.3 defines foreign exchange rates and arbitrage, and examines the relationshipbetween the exchange rate and the nation’s balance of payments. Section 14.4defines spot and forward rates and discusses foreign exchange swaps, futures, andoptions. Section 14.5 then deals with foreign exchange risks, hedging, and spec-ulation. Section 14.6 examines uncovered and covered interest arbitrage, as wellas the efficiency of the foreign exchange market. Finally, Section 14.7 deals withthe Eurocurrency, Eurobond, and Euronote markets. In the appendix, we derivethe formula for the precise calculation of the covered interest arbitrage margin.

14.2 Functions of the Foreign Exchange MarketsBy far the principal function of foreign exchange markets is the transfer of funds orpurchasing power from one nation and currency to another. This is usually accom-plished by an electronic transfer and increasingly through the Internet. With it,

423


424 Foreign Exchange Markets and Exchange Rates

a domestic bank instructs its correspondent bank in a foreign monetary center to pay aspecified amount of the local currency to a person, firm, or account.

The demand for foreign currencies arises when tourists visit another country and need toexchange their national currency for the currency of the country they are visiting, when adomestic firm wants to import from other nations, when an individual or firm wants to investabroad, and so on. Conversely, a nation’s supply of foreign currencies arises from foreigntourist expenditures in the nation, from export earnings, from receiving foreign investments,and so on. For example, suppose a U.S. firm exporting to the United Kingdom is paid inpounds sterling (the U.K. currency). The U.S. exporter will exchange the pounds for dollarsat a commercial bank. The commercial bank will then sell these pounds for dollars to aU.S. resident who is going to visit the United Kingdom, to a U.S. firm that wants to importfrom the United Kingdom and pay in pounds, or to a U.S. investor who wants to invest inthe United Kingdom and needs the pounds to make the investment.

Thus, a nation’s commercial banks operate as clearinghouses for the foreign exchangedemanded and supplied in the course of foreign transactions by the nation’s residents. Inthe absence of this function, a U.S. importer needing British pounds, for instance, wouldhave to locate a U.S. exporter with pounds to sell. This would be very time-consumingand inefficient and would essentially be equivalent to reverting to barter trade. Those U.S.commercial banks that find themselves with an oversupply of pounds will sell their excesspounds (through the intermediary of foreign exchange brokers) to commercial banks thathappen to be short of pounds needed to satisfy their customers’ demand. In the final analysis,then, a nation pays for its tourist expenditures abroad, its imports, its investments abroad,and so on with its foreign exchange earnings from tourism, exports, and the receipt offoreign investments.

If the nation’s total demand for foreign exchange in the course of its foreign transactionsexceeds its total foreign exchange earnings, the rate at which currencies exchange for oneanother will have to change (as explained in the next section) to equilibrate the total quanti-ties demanded and supplied. If such an adjustment in the exchange rates were not allowed,the nation’s commercial banks would have to borrow from the nation’s central bank. Thenation’s central bank would then act as the “lender of last resort” and draw down its for-eign exchange reserves (a balance-of-payments deficit of the nation). On the other hand,if the nation generated an excess supply of foreign exchange in the course of its businesstransactions with other nations (and if adjustment in exchange rates were not allowed), thisexcess supply would be exchanged for the national currency at the nation’s central bank,thus increasing the nation’s foreign currency reserves (a balance-of-payments surplus).

Thus, four levels of transactors or participants can be identified in foreign exchangemarkets. At the bottom, or at the first level, are such traditional users as tourists, importers,exporters, investors, and so on. These are the immediate users and suppliers of foreign cur-rencies. At the next, or second, level are the commercial banks, which act as clearinghousesbetween users and earners of foreign exchange. At the third level are foreign exchange bro-kers, through whom the nation’s commercial banks even out their foreign exchange inflowsand outflows among themselves (the so-called interbank or wholesale market). Finally, atthe fourth and highest level is the nation’s central bank, which acts as the seller or buyer oflast resort when the nation’s total foreign exchange earnings and expenditures are unequal.The central bank then either draws down its foreign exchange reserves or adds to them.


14.2 Functions of the Foreign Exchange Markets 425

Because of the special position of the U.S. dollar as an international currency as well asthe national currency of the United States, U.S. importers and U.S. residents wishing to makeinvestments abroad could pay in dollars. Then it would be U.K. exporters and investmentrecipients who would have to exchange dollars for pounds in the United Kingdom. Similarly,U.S. exporters and U.S. recipients of foreign investments may require payment in dollars.Then it would be U.K. importers or investors who would have to exchange pounds for dollarsin London. This makes foreign monetary centers relatively larger than they otherwise mighthave been.

But the U.S. dollar is more than an international currency. It is a vehicle currency; thatis, the dollar is also used for transactions that do not involve the United States at all, as,for example, when a Brazilian importer uses dollars to pay a Japanese exporter (see CaseStudy 14-1). The same is true of the euro, the newly established currency of the EuropeanMonetary Union or EMU. The United States receives a seignorage benefit when the dollaris used as a vehicle currency. This arises from and amounts to an interest-free loan fromforeigners to the United States on the amount of dollars held abroad. More than 60 percentof the U.S. currency is now held abroad.

The Bank for International Settlements (BIS) in Basel, Switzerland, estimated that thetotal of foreign exchange trading or “turnover” for the world as a whole averaged $4.0trillion per day in 2010, up from $3.3 trillion in 2007, $1.9 trillion in 2004, and $1.2trillion in 2001. This is about 27 percent of the average yearly volume of world trade andof the U.S. gross domestic product (GDP) in 2010. Banks located in the United Kingdom

(continued)

■ CASE STUDY 14-1 The U.S. Dollar as the Dominant International Currency

Today the U.S. dollar is the dominant internationalcurrency, serving as a unit of account, mediumof exchange, and store of value not only fordomestic transactions but also for private andofficial international transactions. The U.S. dollarreplaced the British pound sterling after WorldWar II as the dominant vehicle currency becauseof its more stable value, the existence of largeand well-developed financial markets in theUnited States, and the very large size of the U.S.economy. Since its creation at the beginning of1999, the euro (the common currency of 17 ofthe 27-member countries of the European Union)has become the second most important vehicleinternational currency (see Case Study 14-2).

Table 14.1 shows the relative importance ofthe dollar, the euro, and other major currencies in

the world economy in 2010. The table shows that42.5 percent of foreign exchange trading was indollars, as compared with 19.6 percent in euro, 9.5percent in Japanese yen, and smaller percentagesin other currencies. Table 14.1 also shows that58.2 percent of international bank loans, 38.2percent of international bond offerings, and52.0 percent of international trade invoicing weredenominated in U.S. dollars. Also, 61.5 percent offoreign exchange reserves were held in U.S. dol-lars, as compared with 26.2 percent in euro, andmuch smaller percentages for the yen and othercurrencies. Although the U.S. dollar has graduallylost its role as the sole vehicle currency that itenjoyed since the end of World War II, it stillremains the dominant vehicle currency in the worldtoday.

Salvatore c14.tex V2 - 10/18/2012 1:15 P.M. 26


■ CASE STUDY 14-1 Continued

■ TABLE 14.1. Relative International Importance of Major Currencies in 2010(in Percentages)

Foreign International International ForeignExchange Bank Bond Trade ExchangeTradinga Loansa Offeringa Invoicingb Reservesc

U.S. dollar 42.5 58.2 38.2 52.0 61.5Euro 19.6 21.4 45.1 24.8 26.2Japanese yen 9.5 3.0 3.8 4.7 3.8Pound sterling 6.5 5.5 8.0 5.4 4.0Swiss franc 3.2 2.1 1.5 na 0.1Other currencies 18.7 9.8 4.4 13.1 4.4

aBank of International Settlements, Triennial Central Bank Survey (Basel, Switzerland: BIS, March 2010) andBIS data set.bP. Bekx, ‘‘The Implications of the Introduction of the Euro for Non-EU Countries,’’ Euro Paper No. 26, July1998. Data are for 1995. More recent data are not available.cInternational Monetary Fund, Annual Report (Washington, D.C.: IMF, 2011).

accounted for nearly 37 percent of all foreign exchange market turnover, followed by theUnited States with about 18 percent, Japan with about 6 percent, Singapore, Switzerland,and Hong Kong SAR each with about 5 percent, Australia with about 4 percent, and the restwith other smaller markets. Most of these foreign exchange transactions take place throughdebiting and crediting bank accounts rather than through actual currency exchanges. Forexample, a U.S. importer will pay for EMU goods by debiting his or her account at a U.S.bank. The latter will then instruct its correspondent bank in an EMU country to credit theaccount of the EMU exporter with the euro value of the goods.

Another function of foreign exchange markets is the credit function. Credit is usuallyneeded when goods are in transit and also to allow the buyer time to resell the goods andmake the payment. In general, exporters allow 90 days for the importer to pay. However,the exporter usually discounts the importer’s obligation to pay at the foreign departmentof his or her commercial bank. As a result, the exporter receives payment right away, andthe bank will eventually collect the payment from the importer when due. Still anotherfunction of foreign exchange markets is to provide the facilities for hedging and speculation(discussed in Section 14.5). Today, about 90 percent of foreign exchange trading reflectspurely financial transactions and only about 10 percent trade financing.

With electronic transfers, foreign exchange markets have become truly global in the sensethat currency transactions now require only a few seconds to execute and can take place 24hours per day. As banks end their regular business day in San Francisco and Los Angeles,they open in Singapore, Hong Kong, Sydney, and Tokyo; by the time the latter banks winddown their regular business day, banks open in London, Paris, Zurich, Frankfurt, and Milan;and before the latter close, New York and Chicago banks open.

Case Study 14-1 examines the U.S. dollar as the dominant vehicle currency, whereasCase Study 14-2 discusses the birth of the euro, which has quickly become the second mostimportant vehicle currency.


14.3 Foreign Exchange Rates 427

■ CASE STUDY 14-2 The Birth of a New Currency: The Euro

On January 1, 1999, the euro (¤) came into ex-istence as the single currency of 11 of the then 15member countries of the European Union (Austria,Belgium, Germany, Finland, France, Ireland, Italy,Luxembourg, Spain, Portugal, and the Nether-lands). Greece was admitted at the beginning of2001, Slovenia in 2007, Cyprus and Malta in 2008,Slovakia in 2009, and Estonia in 2011—makingthe number of EMU countries in the Eurozoneequal to 17 (out of the 27 members of the EuropeanUnion or EU in 2011). Britain, Sweden, and Den-mark chose not to participate, but reserved the rightto join later. This was the first time that a groupof sovereign nations voluntarily gave up their cur-rency in favor of a common currency, and it ranksas one of the most important economic events ofthe postwar period.

From the start, the euro became an impor-tant international currency because the EuropeanMonetary Union or EMU (1) is as large an eco-nomic and trading unit as the United States; (2)has a large, well-developed, and growing finan-cial market, which is increasingly free of controls;and (3) has a good inflation performance that willkeep the value of the euro stable. But it is notlikely that the euro will displace the U.S. dollaras the leading international or vehicle currencyany time soon because (1) most primary com-modities are priced in dollars, and this is likely toremain the case for some time to come; (2) mostnon-EMU countries are likely to continue to use

the dollar for most of their international transac-tions for the foreseeable future, with the exceptionof the former communist nations in Central andEastern Europe (which are candidates for admis-sion into the European Monetary Union and mayeven adopt the euro before then) and the formerFrench colonies in West and Central Africa; and(3) sheer inertia favors the incumbent (thedollar).

The most likely situation will be that theeuro will share the leading position with the dollarduring this decade and also with the renminbi oryuan, the currency of China, after that. Althoughstill officially inconvertible, China has alreadystarted rapidly “internationalizing” its currencyby developing an offshore market in the cur-rency and encouraging the use of renminbi insettling and invoicing international trade transac-tions. The World Bank predicted that by 2025 theeuro and the renminbi will become as importantinternational or vehicle currencies as the dollarin a new “multi-currency” international monetarysystem.

Sources: D. Salvatore, “The Euro: Expectations andPerformance,” Eastern Economic Journal , Winter 2002,pp. 121–136; D. Salvatore, “Euro,” Princeton Encyclopediaof the World Economy (Princeton, N.J.: Princeton Univer-sity Press, 2008), pp. 350–352; World Bank, Multipolarity:The New Global Economy (Washington, D.C., 2011), pp.139–142; and D. Salvatore, “Exchange Rate Misalignmentsand the International Monetary System,” Journal of PolicyModeling , July/August 2012, pp. 594–604.

14.3 Foreign Exchange RatesIn this section, we first define exchange rates and show how they are determined under aflexible exchange rate system. Then we explain how exchange rates between currencies areequalized by arbitrage among different monetary centers. Finally, we show the relationshipbetween the exchange rate and the nation’s balance of payments.

14.3A Equilibrium Foreign Exchange RatesAssume for simplicity that there are only two economies, the United States and the EuropeanMonetary Union (EMU), with the dollar ($) as the domestic currency and the euro (¤) as



0 50 100 150 200 250 300 350

0.50

1.00

1.50

2.00

R= $/

A

B

F

G

E

H C

Million /day

D

S

FIGURE 14.1. The Exchange Rate under a Flexible Exchange Rate System.The vertical axis measures the dollar price of the euro (R = $/¤), and the horizontal axis mea-sures the quantity of euros. With a flexible exchange rate system, the equilibrium exchangerate is R = 1, at which the quantity demanded and the quantity supplied are equal at ¤200 mil-lion per day. This is given by the intersection at point E of the U.S. demand and sup-ply curves for euros. At a higher exchange rate, a surplus of euros would result that wouldtend to lower the exchange rate toward the equilibrium rate. At an exchange rate lower thanR = 1, a shortage of euros would result that would drive the exchange rate up toward the equilibriumlevel.

the foreign currency. The exchange rate (R) between the dollar and the euro is equal to thenumber of dollars needed to purchase one euro. That is, R = $/¤. For example, if R = $/¤= 1, this means that one dollar is required to purchase one euro.

Under a flexible exchange rate system of the type we have today, the dollar price ofthe euro (R) is determined, just like the price of any commodity, by the intersection ofthe market demand and supply curves for euros. This is shown in Figure 14.1, where thevertical axis measures the dollar price of the euro, or the exchange rate, R = $/¤, and thehorizontal axis measures the quantity of euros. The market demand and supply curves foreuros intersect at point E , defining the equilibrium exchange rate of R = 1, at which thequantity of euros demanded and the quantity supplied are equal at ¤200 million per day.At a higher exchange rate, the quantity of euros supplied exceeds the quantity demanded,and the exchange rate will fall toward the equilibrium rate of R = 1. At an exchange ratelower than R = 1, the quantity of euros demanded exceeds the quantity supplied, and theexchange rate will be bid up toward the equilibrium rate of R = 1. If the exchange ratewere not allowed to rise to its equilibrium level (as under the fixed exchange rate systemthat prevailed until March 1973), then either restrictions would have to be imposed on thedemand for euros of U.S. residents or the U.S. central bank (the Federal Reserve System)would have to fill or satisfy the excess demand for euros out of its international reserves.



The U.S. demand for euros is negatively inclined, indicating that the lower the exchangerate (R), the greater the quantity of euros demanded by U.S. residents. The reason is thatthe lower the exchange rate (i.e., the fewer the number of dollars required to purchase aeuro), the cheaper it is for U.S. residents to import from and to invest in the EuropeanMonetary Union, and thus the greater the quantity of euros demanded by U.S. residents.On the other hand, the U.S. supply of euros is usually positively inclined (see Figure 14.1),indicating that the higher the exchange rate (R), the greater the quantity of euros earnedby U.S. residents and supplied to the United States. The reason is that at higher exchangerates, EMU residents receive more dollars for each of their euros. As a result, they find U.S.goods and investments cheaper and more attractive and spend more in the United States,thus supplying more euros to the United States.

If the U.S. demand curve for euros shifted up (for example, as a result of increasedU.S. tastes for EMU goods) and intersected the U.S. supply curve for euros at point G (seeFigure 14.1), the equilibrium exchange rate would be R = 1.50, and the equilibrium quantityof euros would be ¤300 million per day. The dollar is then said to have depreciated sinceit now requires $1.50 (instead of the previous $1) to purchase one euro. Depreciation thusrefers to an increase in the domestic price of the foreign currency. Conversely, if the U.S.demand curve for euros shifted down so as to intersect the U.S. supply curve for euros atpoint H (see Figure 14.1), the equilibrium exchange rate would fall to R = 0.5 and thedollar is said to have appreciated (because fewer dollars are now required to purchase oneeuro). Appreciation thus refers to a decline in the domestic price of the foreign currency.An appreciation of the domestic currency means a depreciation of the foreign currency andvice versa. Shifts in the U.S. supply curve for euros would similarly affect the equilibriumexchange rate and equilibrium quantity of euros (these are left as end-of-chapter problems).

The exchange rate could also be defined as the foreign currency price of a unit of thedomestic currency. This is the inverse, or reciprocal, of our previous definition. Since in thecase we examined previously, the dollar price of the euro is R = 1, its inverse is also 1. Ifthe dollar price of the euro were instead R = 2, then the euro price of the dollar would be1/R = 1/2, or it would take half a euro to purchase one dollar. Although this definition ofthe exchange rate is sometimes used, we will use the previous one, or the dollar price ofthe euro (R), unless clearly stated to the contrary. In the real world, the particular definitionof the exchange rate being used is generally spelled out to avoid confusion (see CaseStudy 14-3).

Finally, while we have dealt with only two currencies for simplicity, in reality there arenumerous exchange rates, one between any pair of currencies. Thus, besides the exchangerate between the U.S. dollar and the euro, there is an exchange rate between the U.S. dollarand the British pound (£), between the U.S. dollar and the Swiss franc, the Canadian dollarand the Mexican peso, the British pound and the euro, the euro and the Swiss franc, andbetween each of these currencies and the Japanese yen. Once the exchange rate between eachof a pair of currencies with respect to the dollar is established, however, the exchange ratebetween the two currencies themselves, or cross-exchange rate, can easily be determined.For example, if the exchange rate (R) were 2 between the U.S. dollar and the British poundand 1.25 between the dollar and the euro, then the exchange rate between the pound andthe euro would be 1.60 (i.e., it takes ¤1.6 to purchase 1£). Specifically,

R = ¤/£ = $ value of £

$ value of ¤= 2

1.25= 1.60



■ CASE STUDY 14-3 Foreign Exchange Quotations

Table 14.2 gives the exchange or spot rate for var-ious currencies with respect to the U.S. dollar forFriday May 25, 2012—defined first as the dollarprice of the foreign currency (often referred to asin direct or “American” terms) and then as theforeign currency price of the dollar (i.e., in indi-rect or “European” terms). For example, next tothe Euro area, we find that the direct spot rate was$1.2518/¤1. On the same line, we find that theindirect or euro price of the dollar was ¤0.7988/$1.The last column of the table, headed “U.S. $ vs.

■ TABLE 14.2. Foreign Exchange Quotation, May 25, 2012

CurrenciesU.S.-dollar foreign-exchange rates in late New York trading

US$ vs,—Thurs— YTD chg

Country/currency in US$ per US$ (%)AmericasArgentina peso∗ .2239 4.4668 3.7Brazil real .5031 1.9877 6.5Canada dollar .9716 1.0293 0.8Chile peso .001965 509.00 −2.1Colombia peso .0005462 1831.00 −5.6Ecuador US dollar 1 1 unchMexico peso∗ .0713 14.0238 0.6Peru new sol .3711 2.695 −0.1Uruguay peso† .04975 20.1005 1.5Venezuela b.fuerte .229885 4.3500 unch

Asia-PacificAustralian dollar .9759 1.0247 4.6

1-mos forward .9730 1.0278 4.33-mos forward .9681 1.0330 4.26-mos forward .9618 1.0398 4.1

China yuan .1578 6.3372 0.3Hong Kong dollar .1288 7.7634 unchIndia rupee .01806 55.375 4.4Indonesia rupiah .0001058 9450 4.6Japan yen .012549 79.68 3.6

1-mos forward .012552 79.67 3.53-mos forward .012561 79.61 3.66-mos forward .012581 79.48 3.6

Malaysia ringglt .3171 3.1532 −0.8New Zealand dollar .7537 1.3267 3.2Pakistan rupee .01086 92.055 2.5Philippines peso .0228 43.770 −0.2Singapore dollar .7804 1.2814 −1.2South Korea won .0008432 1185.90 2.2Taiwan dollar .03374 29.640 −2.1Thailand baht .03157 31.676 0.2Vietnam dong .00004796 20850 −0.9

US$ vs,—Thurs— YTD chg

Country/currency in US$ per US$ (%)EuropeCzech Rep. koruna∗∗ .04933 20.273 2.6Denmark krone .1685 5.9355 3.5Euro area euro 1.2518 .7988 3.5Hungary forint .004176 239.44 −1.5Norway krone .1662 6.0162 0.7Poland zloty .2868 3.4869 1.2Russia ruble‡ .03119 32.064 −0.3Sweden krona .1393 7.1766 4.3Switzerland franc 1.0422 .9595 2.4

1-mos forward 1.0427 .9590 2.33-mos forward 1.0443 .9576 2.36-mos forward 1.0472 .9549 2.3

Turkey lira∗∗ .5407 1.8494 −3.5UK pound 1.5660 .6386 −0.8

1-mos forward 1.5657 .6387 −0.83-mos forward 1.5652 .6389 −0.86-mos forward 1.5647 .6391 −0.9

Middle East/AfricaBahrain dinar 2.6532 .3769 unchEgypt pound∗ .1656 6.0372 −0.2Israel shekel .2593 3.8559 1.2Jordan dinar 1.4119 .7083 −0.2Kuwait dinar 3.5677 .2803 0.8Lebanon pound .0006651 1503.45 −0.1Saudi Arabia riyal .2666 3.7509 unchSouth Africa rand .1190 8.4028 3.9UAE dirham .2723 3.6730 unch

Source : ICAPplc.

*Floating rate † Financial §Government rate ‡Russian Central Bank rate **Commercial rateSource: Reprinted by permission of the Wall Street Journal , @ 2012 Dow Jones & Company, Inc. All rights reserved.

YTD chg (%),” shows the percentage change in theexchange rate, year to date (YTD)—that is, fromthe beginning of the year. For example, the tableshows that the dollar appreciated by 3.5 percentvis-a-vis the euro from the beginning of 2012 toMay 25, 2012. Note that the main exchange ratetable also gives the one-month, three-month, andsix-month forward rate for the Australian dollar,the Japanese yen, the Swiss franc, and the Britishpound. These are discussed in Section 14.4A.



Since over time a currency can depreciate with respect to some currencies and appreciateagainst others, an effective exchange rate is calculated. This is a weighted average of theexchange rates between the domestic currency and that of the nation’s most important tradepartners, with weights given by the relative importance of the nation’s trade with eachof these trade partners (see Section 14.5a). Finally, we must also distinguish between thenominal exchange rate (the one we have been discussing) and the real exchange rate (to bediscussed in Chapter 15).

14.3B ArbitrageThe exchange rate between any two currencies is kept the same in different monetary centersby arbitrage. This refers to the purchase of a currency in the monetary center where it ischeaper, for immediate resale in the monetary center where it is more expensive, in orderto make a profit.

For example, if the dollar price of the euro was $0.99 in New York and $1.01 in Frankfurt,an arbitrageur (usually a foreign exchange dealer of a commercial bank) would purchaseeuros at $0.99 in New York and immediately resell them in Frankfurt for $1.01, thus realizinga profit of $0.02 per euro. While the profit per euro transferred seems small, on ¤1 millionthe profit would be $20,000 for only a few minutes work. From this profit must be deductedthe cost of the electronic transfer and the other costs associated with arbitrage. Since thesecosts are very small, we shall ignore them here.

As arbitrage takes place, however, the exchange rate between the two currencies tendsto be equalized in the two monetary centers. Continuing our example, we see that arbitrageincreases the demand for euros in New York, thereby exerting an upward pressure on thedollar price of euros in New York. At the same time, the sale of euros in Frankfurt increasesthe supply of euros there, thus exerting a downward pressure on the dollar price of euros inFrankfurt. This continues until the dollar price of the euro quickly becomes equal in NewYork and Frankfurt (say at $1 = ¤1), thus eliminating the profitability of further arbitrage.

When only two currencies and two monetary centers are involved in arbitrage, as in thepreceding example, we have two-point arbitrage. When three currencies and three monetarycenters are involved, we have triangular , or three-point, arbitrage. While triangular arbitrageis not very common, it operates in the same manner to ensure consistent indirect , or cross,exchange rates between the three currencies in the three monetary centers. For example,suppose exchange rates are as follows:

$1 = ¤1 in New York

¤1 = £0.64 in Franfurt

£0.64 = $1 in London

These cross rates are consistent because

$1 = ¤1 = £0.64

and there is no possibility of profitable arbitrage. However, if the dollar price of the eurowere $0.96 in New York, with the other exchange rates as indicated previously, then itwould pay to use $0.96 to purchase ¤1 in New York, use the ¤1 to buy £0.64 in Frankfurt,and exchange the £0.64 for $1 in London, thus realizing a $0.04 profit on each euro so



transferred. On the other hand, if the dollar price of the euro was $1.04 in New York, itwould pay to do just the opposite—that is, use $1 to purchase £0.64 in London, exchangethe £0.64 for ¤1 in Frankfurt, and exchange the ¤1 for $1.04 in New York, thus making aprofit of $0.04 on each euro so transferred.

As in the case of two-point arbitrage, triangular arbitrage increases the demand for thecurrency in the monetary center where the currency is cheaper, increases the supply ofthe currency in the monetary center where the currency is more expensive, and quicklyeliminates inconsistent cross rates and the profitability of further arbitrage. As a result,arbitrage quickly equalizes exchange rates for each pair of currencies and results in consistentcross rates among all pairs of currencies, thus unifying all international monetary centersinto a single market.

14.3C The Exchange Rate and the Balance of PaymentsWe can examine the relationship between the exchange rate and the nation’s balance ofpayments with Figure 14.2, which is identical to Figure 14.1 except for the addition ofthe new demand curve for euros labeled D ′

¤. We have seen in Chapter 13 that the U.S.demand for euros (D¤) arises from the U.S. demand for imports of goods and services fromthe European Union, from U.S. unilateral transfers to the European Union, and from U.S.

50 100 200 250 300 350 450

0.50

1.00

1.50

2.00

1.25

0

R = $/

E'

D'

W Z

TE

Million /day

S

D

FIGURE 14.2. Disequilibrium under a Fixed and a Flexible Exchange Rate System.With D¤ and S¤, equilibrium is at point E at the exchange rate of R = $/¤ = 1, at which the quantities

of euros demanded and supplied are equal at ¤200 million per day. If D¤ shifted up to D ′¤, the United

States could maintain the exchange rate at R = 1 by satisfying (out of its official euro reserves) the excessdemand of ¤250 million per day (TE in the figure). With a freely flexible exchange rate system, the dollarwould depreciate until R = 1.50 (point E ′ in the figure). If, on the other hand, the United States wanted tolimit the depreciation of the dollar to R = 1.25 under a managed float, it would have to satisfy the excessdemand of ¤100 million per day (WZ in the figure) out of its official euro reserves.



investments in the European Monetary Union (a capital outflow from the United States).These are the autonomous debit transactions of the United States that involve payments tothe European Monetary Union.

On the other hand, the supply of euros (S¤) arises from U.S. exports of goods and servicesto the European Monetary Union, from unilateral transfers received from the EuropeanMonetary Union, and from the EMU investments in the United States (a capital inflow tothe United States). These are the autonomous credit transactions of the United States thatinvolve payments from the European Monetary Union. (We are assuming for simplicity thatthe United States and the European Monetary Union are the only two economies in theworld and that all transactions between them take place in euros.)

With D¤ and S¤, the equilibrium exchange rate is R = $/¤ = 1 (point E in Figure 14.2),at which ¤200 million are demanded and supplied per day (exactly as in Figure 14.1). Nowsuppose that for whatever reason (such as an increase in U.S. tastes for EMU products)the U.S. autonomous demand for euros shifts up to D ′

¤. If the United States wanted tomaintain the exchange rate fixed at R = 1, U.S. monetary authorities would have to satisfythe excess demand for euros of TE (¤250 million per day in Figure 14.2) out of its officialreserve holdings of euros. Alternatively, EMU monetary authorities would have to purchasedollars (thus adding to their official dollar reserves) and supply euros to the foreign exchangemarket to prevent an appreciation of the euro (a depreciation of the dollar). In either case,the U.S. official settlements balance would show a deficit of ¤250 million ($250 million atthe official exchange rate of R = 1) per day, or ¤91.25 billion ($91.25 billion) per year.

If, however, the United States operated under a freely flexible exchange rate system, theexchange rate would rise (i.e., the dollar would depreciate) from R = 1.00 to R = 1.50, atwhich the quantity of euros demanded (¤300 million per day) exactly equals the quantitysupplied (point E ′ in Figure 14.2). In this case, the United States would not lose any of itsofficial euro reserves. Indeed, international reserves would be entirely unnecessary undersuch a system. The tendency for an excess demand for euros on autonomous transactionswould be completely eliminated by a sufficient depreciation of the dollar with respect tothe euro.

However, under a managed floating exchange rate system of the type in operation since1973, U.S. monetary authorities can intervene in foreign exchange markets to moderatethe depreciation (or appreciation) of the dollar. In the preceding example, the United Statesmight limit the depreciation of the dollar to R = 1.25 (instead of letting the dollar depreciateall the way to R = 1.50 as under a freely fluctuating exchange rate system). The UnitedStates could do this by supplying to the foreign exchange market the excess demand foreuros of WZ , or ¤100 million per day, out of its official euro reserves (see the figure). Undersuch a system, part of the potential deficit in the U.S. balance of payments is covered bythe loss of official reserve assets of the United States, and part is reflected in the form of adepreciation of the dollar. Thus, we cannot now measure the deficit in the U.S. balance ofpayments by simply measuring the loss of U.S. international reserves or by the amount ofthe net credit balance in the official reserve account of the United States. Under a managedfloat, the loss of official reserves only indicates the degree of official intervention in foreignexchange markets to influence the level and movement of exchange rates, and not thebalance-of-payments deficit.

For this reason, since 1976 the United States has suspended the calculation of thebalance-of-payments deficit or surplus. The statement of international transactions does



not even show the net balance on the official reserve account (although it can be easilycalculated) in order to be neutral and not to focus undue attention on such a balance, inview of the present system of floating but managed exchange rates (see Table 13.1).

The concept and measurement of international transactions and the balance of paymentsare still very important and useful, however, for several reasons. First, as pointed out inChapter 13, the flow of trade provides the link between international transactions and thenational income. (This link is examined in detail in Chapter 17.) Second, many developingcountries still operate under a fixed exchange rate system and peg their currency to a majorcurrency, such as the U.S. dollar and the euro, or to SDRs. Third, the International MonetaryFund requires all member nations to report their balance-of-payments statement annuallyto it (in the specific format shown in Section A13.1). Finally, and perhaps more important,while not measuring the deficit or surplus in the balance of payments, the balance of theofficial reserve account gives an indication of the degree of intervention by the nation’smonetary authorities in the foreign exchange market to reduce exchange rate volatility andto influence exchange rate levels.

14.4 Spot and Forward Rates, Currency Swaps,Futures, and Options

In this section we distinguish between spot and forward exchange rates and examine theirsignificance. Then we discuss foreign exchange swaps, futures, and options and their uses.

14.4A Spot and Forward RatesThe most common type of foreign exchange transaction involves the payment and receipt ofthe foreign exchange within two bussiness days after the day the transaction is agreed upon.The two-day period gives adequate time for the parties to send instructions to debit andcredit the appropriate bank accounts at home and abroad. This type of transaction is calleda spot transaction , and the exchange rate at which the transaction takes place is called thespot rate. The exchange rate R = $/¤ = 1 in Figure 14.1 is a spot rate.

Besides spot transactions, there are forward transactions. A forward transaction involvesan agreement today to buy or sell a specified amount of a foreign currency at a specifiedfuture date at a rate agreed upon today (the forward rate). For example, I could enter intoan agreement today to purchase ¤100 three months from today at $1.01 = ¤1. Note that nocurrencies are paid out at the time the contract is signed (except for the usual 10 percentsecurity margin). After three months, I get the ¤100 for $101, regardless of what the spotrate is at that time. The typical forward contract is for one month, three months, or sixmonths, with three months the most common (see Case Study 14-3). Forward contracts forlonger periods are not as common because of the great uncertainties involved. However,forward contracts can be renegotiated for one or more periods when they become due. Inwhat follows, we will deal exclusively with three-month forward contracts and rates, butthe procedure would be the same for forward contracts of different duration.

The equilibrium forward rate is determined at the intersection of the market demandand supply curves of foreign exchange for future delivery . The demand for and supply offorward foreign exchange arise in the course of hedging, from foreign exchange speculation,


14.4 Spot and Forward Rates, Currency Swaps, Futures, and Options 435

and from covered interest arbitrage. These, as well as the close relationship between thespot rate and the forward rate, are discussed next in Sections 14.5 and 14.6. All that needsto be said here is that, at any point in time, the forward rate can be equal to, above, orbelow the corresponding spot rate.

If the forward rate is below the present spot rate, the foreign currency is said to be ata forward discount with respect to the domestic currency. However, if the forward rate isabove the present spot rate, the foreign currency is said to be at a forward premium. Forexample, if the spot rate is $1 = ¤1 and the three-month forward rate is $0.99 = ¤1, we saythat the euro is at a three-month forward discount of 1 cent or 1 percent (or at a 4 percentforward discount per year) with respect to the dollar. On the other hand, if the spot rate isstill $1 = ¤1 but the three-month forward rate is instead $1.01 = ¤1, the euro is said to beat a forward premium of 1 cent or 1 percent for three months, or 4 percent per year.

Forward discounts (FD) or premiums (FP) are usually expressed as percentages per yearfrom the corresponding spot rate and can be calculated formally with the following formula:

FD or FP = FR − SR

SR× 4 × 100

where FR is the forward rate and SR is the spot rate (what we simply called R in theprevious section). The multiplication by 4 is to express the FD(−) or FP (+) on a yearlybasis, and the multiplication by 100 is to express the FD or FP in percentages. Thus, whenthe spot rate of the pound is SR = $1.00 and the forward rate is FR = $0.99, we get

FD = $0.99 − $1.00

$1.00× 4 × 100 = −$0.01

$1.00× 4 × 100

= −0.01 × 4 × 100 = −4%

the same as found earlier without the formula. Similarly, if SR = $1 and FR = $1.01:

FP = $1.01 − $1.00

$1.00× 4 × 100 = $0.01

$1.00× 4 × 100

= 0.01 × 4 × 100 = +4%

14.4B Foreign Exchange SwapsA foreign exchange swap refers to a spot sale of a currency combined with a forwardrepurchase of the same currency—as part of a single transaction. For example, supposethat Citibank receives a $1 million payment today that it will need in three months, but inthe meantime it wants to invest this sum in euros. Citibank would incur lower brokeragefees by swapping the $1 million into euros with Frankfurt’s Deutsche Bank as part of asingle transaction or deal, instead of selling dollars for euros in the spot market today andat the same time repurchasing dollars for euros in the forward market for delivery in threemonths—in two separate transactions. The swap rate (usually expressed on a yearly basis)is the difference between the spot and forward rates in the currency swap.

Most interbank trading involving the purchase or sale of currencies for future deliveryis done not by forward exchange contracts alone but combined with spot transactions in



the form of foreign exchange swaps. In April 2010, there were $1,765 billion worth of for-eign exchange swaps outstanding. These represented 44 percent of total interbank currencytrading. Spot transactions were $1,490 billion or 37 percent of the total. Thus, the foreignexchange market is dominated by the foreign exchange swap and spot markets.

14.4C Foreign Exchange Futures and OptionsAn individual, firm, or bank can also purchase or sell foreign exchange futures and options.Trading in foreign exchange futures was initiated in 1972 by the International MonetaryMarket (IMM) of the Chicago Mercantile Exchange (CME). A foreign exchange futures isa forward contract for standardized currency amounts and selected calendar dates traded onan organized market (exchange). The currencies traded on the IMM are the Japanese yen,the Canadian dollar, the British pound, the Swiss franc, the Australian dollar, the Mexicanpeso, and the euro.

International Monetary Market trading is done as contracts of standard size. Forexample, the IMM Japanese yen contract is for ¥12.5 million, the Canadian dollarcontract is for C$100,000, the pound contract is for £62,500, and the euro contract isfor ¤125,000. Only four dates per year are available: the third Wednesday in March,June, September, and December (see Case Study 14-4). The IMM imposes a dailylimit on exchange rate fluctuations. Buyers and sellers pay a brokerage commission andare required to post a security deposit or margin (about 4 percent of the value of thecontract). A market similar to the IMM is the NYSE Euronext Liffe and the Frankfurt-basedEurex.

The futures market differs from a forward market in that in the futures market only a fewcurrencies are traded; trades occur in standardized contracts only, for a few specific deliverydates, and are subject to daily limits on exchange rate fluctuations; and trading takes placeonly in a few geographical locations, such as Chicago, New York, London, Frankfurt, andSingapore. Futures contracts are usually for smaller amounts than forward contracts and thusare more useful to small firms than to large ones but are somewhat more expensive. Futurescontracts can also be sold at any time up until maturity on an organized futures market,while forward contracts cannot. While the market for currency futures is small comparedwith the forward market, it has grown very rapidly, especially in recent years. (The valueof currency futures outstanding was about $475 billion in April 2010). The two markets arealso connected by arbitrage when prices differ.

Since 1982, individuals, firms, and banks have also been able to buy foreign exchangeoptions (in Japanese yen, Canadian dollars, British pounds, Swiss francs, and euros) on thePhiladelphia Stock Exchange, the Chicago Mercantile Exchange (since 1984), or from abank. A foreign exchange option is a contract giving the purchaser the right, but not theobligation, to buy (a call option) or to sell (a put option) a standard amount of a tradedcurrency on a stated date (the European option) or at any time before a stated date (theAmerican option) and at a stated price (the strike or exercise price). Foreign exchangeoptions are in standard sizes equal to those of futures IMM contracts. The buyer of theoption has the choice to purchase or forego the purchase if it turns out to be unprofitable.The seller of the option, however, must fulfill the contract if the buyer so desires. The buyerpays the seller a premium (the option price) ranging from 1 to 5 percent of the contract’svalue for this privilege when he or she enters the contract. About $207 billion of currencyoptions were outstanding in April 2010.


14.4 Spot and Forward Rates, Currency Swaps, Futures, and Options 437

■ CASE STUDY 14-4 Size, Currency, and Geographic Distributionof the Foreign Exchange Market

Table 14.3 gives data on the size, currency, andgeographical distribution of the foreign exchangemarket in 2010. The table shows that averagedaily spot transactions amounted to $1,490 bil-lion or 37.4 percent of the total market turnover,outright forwards (forward transactions or futures)were $475 billion or 11.9 percent of the total, for-eign exchange swaps were $1,765 billion or 44.3percent, currency swaps (foreign exchange deriva-tives) were $43 billion or 1.1 percent, and optionsand other products were $207 billion or 5.2 percent,

■ TABLE 14.3. Average Daily Global Foreign Exchange Market Turnover, Currency,and Geographic Distribution in 2010

Market Turnovera Currency Distribution Geographic Distribution

Value % of(billion $) Total Currency % Shareb Nation % Share

Spot transactions 1, 490 37.4 U.S. dollar 84.9 United Kingdom 36.7Outright forwards 475 11.9 Euro 39.1 United States 17.9Foreign exchange swaps 1, 765 44.3 Japanese yen 19.0 Japan 6.2Currency swaps 43 1.1 British pound 12.9 Singapore 5.3Options and other products 207 5.2 Australian dollar 7.6 Switzerland 5.2Total 3, 981 100.0 Swiss franc 6.4 Hong Kong 4.7

Canadian dollar 5.3 Australia 3.8Hong Kong dollar 2.4 France 3.0Other 22.4 Other 17.2Total 200.0 Total 100.0

aDaily averages in April, in billions of U.S. dollars; total does not add up because of rounding.bTotal market shares sum to 200 percent rather than to 100 percent because each transaction involves two currencies.Source: Bank for International Settlements, Triennial Central Bank Survey (Basel: BIS), December 2010.

for a grand total foreign exchange market of $3,981billion in 2010. The table also shows that the shareof the U.S. dollar was more than twice that of theeuro and more than four that of the Japanese yenand more than six times that of the British pound(the two currencies most used after the dollar andthe euro). The United Kingdom (mostly London)had the largest share of the market with 36.7 per-cent followed by the United States (mostly NewYork, Chicago, and Philadelphia) with 17.9 percentshare.

In contrast, neither forward contracts nor futures are options. Although forward contractscan be reversed (e.g., a party can sell a currency forward to neutralize a previous purchase)and futures contracts can be sold back to the futures exchange, both must be exercised (i.e.,both contracts must be honored by both parties on the delivery date). Thus, options are lessflexible than forward contracts, but in some cases they may be more useful. For example,an American firm making a bid to take over an EMU firm may be required to promise topay a specified amount in euros. Since the American firm does not know if its bid will besuccessful, it will purchase an option to buy the euros that it would need and will exercisethe option if the bid is successful. Case Study 14-4 gives the average daily distribution ofglobal foreign exchange market turnover by instrument, by currency, and by geographicallocation.



14.5 Foreign Exchange Risks, Hedging, and SpeculationIn this section, we examine the meaning of foreign exchange risks and how they can beavoided or covered by individuals and firms whose main business is not speculation. Wethen discuss how speculators attempt to earn a profit by trying to anticipate future foreignexchange rates.

14.5A Foreign Exchange RisksThrough time, a nation’s demand and supply curves for foreign exchange shift, causing thespot (and the forward) rate to vary frequently. A nation’s demand and supply curves forforeign exchange shift over time as a result of changes in tastes for domestic and foreignproducts in the nation and abroad, different growth and inflation rates in different nations,changes in relative rates of interest, changing expectations, and so on.

For example, if U.S. tastes for EMU products increase, the U.S. demand for eurosincreases (the demand curve shifts up), leading to a rise in the exchange rate (i.e., a depreci-ation of the dollar). On the other hand, a lower rate of inflation in the United States than inthe European Monetary Union leads to U.S. products becoming cheaper for EMU residents.This tends to increase the U.S. supply of euros (the supply curve shifts to the right) andcauses a decline in the exchange rate (i.e., an appreciation of the dollar). Or simply theexpectation of a stronger dollar may lead to an appreciation of the dollar. In short, in adynamic and changing world, exchange rates frequently vary, reflecting the constant changein the numerous economic forces simultaneously at work.

Figure 14.3 shows the great variation in exchange rates of the U.S. dollar with respect tothe Japanese yen, the euro, the British pound, and the Canadian dollar from 1971 to 2005.Note that the exchange rate is here defined from the foreign nation’s point of view (i.e., itis the foreign-currency price of the U.S. dollar), so that an increase in the exchange raterefers to a depreciation of the foreign currency (it takes more units of the foreign currencyto purchase one dollar), while a reduction in the exchange rate refers to an appreciation ofthe foreign currency (and depreciation of the dollar).

The first panel of Figure 14.3 shows the sharp appreciation of the Japanese yen withrespect to the U.S. dollar from about 360 yen per dollar in 1971 to 180 yen in fall 1978.The yen exchange rate then rose (i.e., the yen depreciated) to 260 yen per dollar in fall1982 and again in spring 1985, but then it declined almost continuously until only slightlyabove 80 yen per dollar in spring of 1995; it stayed in the range 109 to 125 yen per dollarbetween 1996 and 2007, and it averaged 82 yen per dollar in March 2012.

The second panel of Figure 14.3 shows that the euro depreciated sharply from $1.17/¤,the value at which it was introduced on January 1, 1999, to $0.85/¤ in October 2000, butthen it appreciated just as sharply from the beginning of 2002 to reach the high of $1.36/¤in December 2004. The euro then depreciated to an average of $1.25/¤ in 2005, it roseto the all-time peak of $1.58/¤ in July 2008, but then it depreciated to $1.32/¤ in March2012. Note that the euro dollar exchange rate in Figure 14.3 is defined as the dollar price ofthe euro (rather than the other way around, as for the exchange rate of the other currenciesshown in Figure 14.3). Note also the sharp depreciation of the British pound with respect tothe U.S. dollar from 1980 to 1985 (and in 2008) and the sharp appreciation of the Canadiandollar with respect to the U.S. from 2002 to the beginning of 2008 (and depreciation in fall2008, followed by appreciation).


14.5 Foreign Exchange Risks, Hedging, and Speculation 439

400

350

300

250

200

150

100

70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12

758. Japan (Yen per U.S. Dollar)

1.6

1.4

1.2

1.0

0.870 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12

751. European Union (U.S. Dollar per Euro)

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.370 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12

752. United Kingdom (Pound per U.S. Dollar)

1.8

1.6

1.4

1.2

1.0

0.870 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12

753. Canada (Canadian Dollar per U.S. Dollar)

160

140

120

100

80

6070 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12

750. Weighted Average, Exchange Value of U.S. Dollar (Index: March 1973=100)

FIGURE 14.3. The Exchange Rate of Major Currencies and the Dollar Effective Exchange Rate,1970–2012.The top four panels of the figure show the fluctuations of the exchange rate of the Japanese yen, theeuro, the British pound, and the Canadian dollar with respect to the dollar from 1970 to 2009 (the euroonly from its creation at the beginning of 1999). The exchange rate used is the foreign-currency value ofthe dollar (so that an increase in the exchange rate refers to a depreciation of the foreign currency andappreciation of the dollar), except for the euro. The bottom panel shows the effective exchange rate ofthe dollar defined as the weighted average of the foreign-currency value of the dollar, with March 1973 =100. The figure shows the wide fluctuations of exchange rates from 1970 to 2012.Source: The Conference Board, Business Cycle Indicators, April 2012, p. 23.



The bottom panel of Figure 14.3 shows the effective exchange rate of the dollar (definedas the weighted average foreign-currency value of the dollar, with March 1973 = 100). Theindex is useful because the exchange rate between the U.S. dollar and the various currencieschanged by different amounts and sometimes in different directions over time. The sharpdepreciation of the other currencies and appreciation of the dollar from the beginning of1980 until the beginning of 1985, as well as the appreciation of the other currencies anddepreciation of the dollar from the beginning of 1985 until the end of 1987, are clearlyshown in the figure. Although less spectacularly, the effective exchange rate of the dollarhas also fluctuated a great deal since 1987, and it was 75 in March 2012 (see Figure 14.3).

The frequent and relatively large fluctuations in exchange rates shown in Figure 14.3impose foreign exchange risks on all individuals, firms, and banks that have to make orreceive future payments denominated in a foreign currency. For example, suppose a U.S.importer purchases ¤100,000 worth of goods from the European Monetary Union and hasto pay in three months in euros. If the present spot rate of the pound is SR = $1/¤1, thecurrent dollar value of the payment that he or she must make in three months is $100,000.However, in three months the spot rate might change to SR = $1.10/¤1. Then the importerwould have to pay $110,000, or $10,000 more, for the imports. Of course, in three monthsthe spot rate might be SR = $0.90/¤1, in which case the importer would have to pay only$90,000, or $10,000 less than anticipated. However, the importer has enough to worry aboutin the import business without also having to deal with this exchange risk. As a result, theimporter will usually want to insure against an increase in the dollar price of the euro (i.e.,an increase in the spot rate) in three months.

Similarly, a U.S. exporter who expects to receive a payment of ¤100,000 in three monthswill receive only $90,000 (instead of the $100,000 that he or she anticipates at today’s spotrate of SR = $1/¤1) if the spot rate in three months is SR = $0.90/¤1. Once again, thespot rate could be higher in three months than it is today so that the exporter would receivemore than anticipated. However, the exporter, like the importer, will usually want to avoid(at a small cost) the exchange risk that he or she faces. Another example is provided byan investor who buys euros at today’s spot rate in order to invest in three-month EMUtreasury bills paying a higher rate than U.S. treasury bills. However, in three months,when the investor wants to convert euros back into dollars, the spot rate may have fallensufficiently to wipe out most of the extra interest earned on the EMU bills or even producea loss.

These three examples clearly show that whenever a future payment must be made orreceived in a foreign currency, a foreign exchange risk, or a so-called open position, isinvolved because spot exchange rates vary over time. In general, businesspeople are riskaverse and will want to avoid or insure themselves against their foreign exchange risk.(Note that arbitrage does not involve any exchange risk since the currency is bought atthe cheaper price in one monetary center to be resold immediately at the higher pricein another monetary center.) A foreign exchange risk arises not only from transactionsinvolving future payments and receipts in a foreign currency (the transaction exposure),but also from the need to value inventories and assets held abroad in terms of the domesticcurrency for inclusion in the firm’s consolidated balance sheet (the translation or accountingexposure), and in estimating the domestic currency value of the future profitability of thefirm (the economic exposure). In what follows, we concentrate on the transaction exposureor risk.



14.5B HedgingHedging refers to the avoidance of a foreign exchange risk, or the covering of an openposition. For example, the importer of the previous example could borrow ¤100,000 at thepresent spot rate of SR = $1/¤1 and leave this sum on deposit in a bank (to earn interest)for three months, when payment is due. By so doing, the importer avoids the risk that thespot rate in three months will be higher than today’s spot rate and that he or she wouldhave to pay more than $100,000 for the imports. The cost of insuring against the foreignexchange risk in this way is the positive difference between the interest rate the importerhas to pay on the loan of ¤100,000 and the lower interest rate he or she earns on thedeposit of ¤100,000. Similarly, the exporter could borrow ¤100,000 today, exchange thissum for $100,000 at today’s spot rate of SR = $1/¤1, and deposit the $100,000 in a bankto earn interest. After three months, the exporter would repay the loan of ¤100,000 withthe payment of ¤100,000 he or she receives. The cost of avoiding the foreign exchange riskin this manner is, once again, equal to the positive difference between the borrowing anddeposit rates of interest.

Covering the foreign exchange risk in the spot market as indicated above has a veryserious disadvantage, however. The businessperson or investor must borrow or tie up his orher own funds for three months. To avoid this, hedging usually takes place in the forwardmarket, where no borrowing or tying up of funds is required. Thus, the importer could buyeuros forward for delivery (and payment) in three months at today’s three-month forwardrate. If the euro is at a three-month forward premium of 4 percent per year, the importerwill have to pay $101,000 in three months for the ¤100,000 needed to pay for the imports.Therefore, the hedging cost will be $1,000 (1 percent of $100,000 for the three months).Similarly, the exporter could sell pounds forward for delivery (and payment) in three monthsat today’s three-month forward rate, in anticipation of receiving the payment of ¤100,000for the exports. Since no transfer of funds takes place until three months have passed, theexporter need not borrow or tie up his or her own funds now. If the euro is at a three-monthforward discount of 4 percent per year, the exporter will get only $99,000 for the ¤100,000he or she delivers in three months. On the other hand, if the euro is at a 4 percent forwardpremium, the exporter will receive $101,000 in three months with certainty by hedging.

A foreign exchange risk can also be hedged and an open position avoided in the futuresor options markets. For example, suppose that an importer knows that he or she must pay¤100,000 in three months and the three-month forward rate of the pound is FR = $1/¤1.The importer could either purchase the ¤100,000 forward (in which case he or she willhave to pay $100,000 in three months and receive the ¤100,000) or purchase an optionto purchase ¤100,000 in three months, say at $1/¤1, and pay now the premium of, say, 1percent (or $1,000 on the $100,000 option). If in three months the spot rate of the poundis SR = $0.98/¤1, the importer would have to pay $100,000 with the forward contract, butcould let the option expire unexercised and get the ¤100,000 at the cost of only $98,000 onthe spot market. In that case, the $1,000 premium can be regarded as an insurance policyand the importer will save $2,000 over the forward contract.

In a world of foreign exchange uncertainty, the ability of traders and investors to hedgegreatly facilitates the international flow of trade and investments. Without hedging therewould be smaller international capital flows, less trade and specialization in production, andsmaller benefits from trade. Note that a large firm, such as a multinational corporation, that



has to make and receive a large number of payments in the same foreign currency at thesame time in the future need only hedge its net open position. Similarly, a bank has an openposition only in the amount of its net balance on contracted future payments and receipts ineach foreign currency at each future date. The bank closes as much of its open positions aspossible by dealing with other banks (through foreign exchange brokers), and it may coverthe remainder in the spot, futures, or options markets.

14.5C SpeculationSpeculation is the opposite of hedging. Whereas a hedger seeks to cover a foreign exchangerisk, a speculator accepts and even seeks out a foreign exchange risk, or an open position,in the hope of making a profit. If the speculator correctly anticipates future changes in spotrates, he or she makes a profit; otherwise, he or she incurs a loss. As in the case of hedging,speculation can take place in the spot, forward, futures, or options markets—usually in theforward market. We begin by examining speculation in the spot market.

If a speculator believes that the spot rate of a particular foreign currency will rise, he orshe can purchase the currency now and hold it on deposit in a bank for resale later. If thespeculator is correct and the spot rate does indeed rise, he or she earns a profit on each unitof the foreign currency equal to the spread between the previous lower spot rate at whichhe or she purchased the foreign currency and the higher subsequent spot rate at which heor she resells it. If the speculator is wrong and the spot rate falls instead, he or she incurs aloss because the foreign currency must be resold at a price lower than the purchase price.

If, on the other hand, the speculator believes that the spot rate will fall, he or she borrowsthe foreign currency for three months, immediately exchanges it for the domestic currencyat the prevailing spot rate, and deposits the domestic currency in a bank to earn interest.After three months, if the spot rate on the foreign currency is lower, as anticipated, thespeculator earns a profit by purchasing the currency (to repay the foreign exchange loan) atthe lower spot rate. (Of course, for the speculator to earn a profit, the new spot rate must besufficiently lower than the previous spot rate to also overcome the possibly higher interestrate paid on a foreign currency deposit over the domestic currency deposit.) If the spot ratein three months is higher rather than lower, the speculator incurs a loss.

In both of the preceding examples, the speculator operated in the spot market and eitherhad to tie up his or her own funds or had to borrow to speculate. It is to avoid this seriousshortcoming that speculation, like hedging, usually takes place in the forward market. Forexample, if the speculator believes that the spot rate of a certain foreign currency will behigher in three months than its present three-month forward rate, the speculator purchasesa specified amount of the foreign currency forward for delivery (and payment) in threemonths. After three months, if the speculator is correct, he or she receives delivery of theforeign currency at the lower agreed forward rate and immediately resells it at the higherspot rate, thus realizing a profit. Of course, if the speculator is wrong and the spot rate inthree months is lower than the agreed forward rate, he or she incurs a loss. In any event, nocurrency changes hands until the three months are over (except for the normal 10 percentsecurity margin that the speculator is required to pay at the time he or she signs the forwardcontract).

As another example, suppose that the three-month forward rate on the euro is FR =$1.01/¤1 and the speculator believes that the spot rate of the euro in three months will be



SR = $0.99/¤1. The speculator then sells euros forward for delivery in three months. Afterthree months, if the speculator is correct and the spot rate is indeed as anticipated, he orshe purchases euros in the spot market at SR = $0.99/¤1 and immediately resells them tofulfill the forward contract at the agreed forward rate of $1.01/¤1, thereby earning a profitof 2 cents per euro. If the spot rate in three months is instead SR = $1.00/¤1, the speculatorearns only 1 cent per euro. If the spot rate in three months is $1.01/¤1, the speculator earnsnothing. Finally, if the spot rate in three months is higher than the forward rate at whichthe speculator sold the forward euros, the speculator incurs a loss on each euro equal to thedifference between the two rates.

As an alternative, the speculator (who believes that the euro will depreciate) could havepurchased an option to sell a specific amount of euros in three months at the rate of,say, $1.01/¤1. If the speculator is correct and the spot rate of the euro in three monthsis indeed $0.99/¤1 as anticipated, he or she will exercise the option, buy euros in thespot market at $0.99/¤1, and receive $1.01/¤1 by exercising the option. By so doing, thespeculator earns 2 cents per euro (from which he or she deducts the premium or the optionprice to determine the net gain). In this case, the result will be the same as with the forwardcontract, except that the option price may exceed the commission on the forward contractso that his or her net profit with the option may be a little less. On the other hand, if thespeculator is wrong and the spot rate of the euro is much higher than expected after threemonths, he or she will let the option contract expire unexercised and incur only the cost ofthe premium or option price. With the forward contract, the speculator would have to honorhis or her commitment and incur a much larger loss.

When a speculator buys a foreign currency on the spot, forward, or futures market, orbuys an option to purchase a foreign currency in the expectation of reselling it at a higherfuture spot rate, he or she is said to take a long position in the currency. On the otherhand, when the speculator borrows or sells forward a foreign currency in the expectation ofbuying it at a future lower price to repay the foreign exchange loan or honor the forwardsale contract or option, the speculator is said to take a short position (i.e., the speculator isnow selling what he or she does not have).

Speculation can be stabilizing or destabilizing. Stabilizing speculation refers to the pur-chase of a foreign currency when the domestic price of the foreign currency (i.e., theexchange rate) falls or is low, in the expectation that it will soon rise, thus leading to aprofit. Or it refers to the sale of the foreign currency when the exchange rate rises or ishigh, in the expectation that it will soon fall. Stabilizing speculation moderates fluctuationsin exchange rates over time and performs a useful function.

On the other hand, destabilizing speculation refers to the sale of a foreign currency whenthe exchange rate falls or is low, in the expectation that it will fall even lower in the future,or the purchase of a foreign currency when the exchange rate is rising or is high, in theexpectation that it will rise even higher in the future. Destabilizing speculation thus magnifiesexchange rate fluctuations over time and can prove very disruptive to the international flowof trade and investments. Whether speculation is primarily stabilizing or destabilizing isa very important question, to which we return in Chapter 16, when we analyze in depththe operation of a flexible exchange rate system, and in Chapter 20, when we comparethe operation of a flexible exchange rate system with that of a fixed exchange rate system.In general, it is believed that under “normal” conditions speculation is stabilizing, and weassume so here.



Speculators are usually wealthy individuals or firms rather than banks. However, anyonewho has to make a payment in a foreign currency in the future can speculate by speeding uppayment if he or she expects the exchange rate to rise and delaying it if he or she expectsthe exchange rate to fall, while anyone who has to receive a future payment in a foreigncurrency can speculate by using the reverse tactics. For example, if an importer expects theexchange rate to rise soon, he or she can anticipate the placing of an order and pay forimports right away. On the other hand, an exporter who expects the exchange rate to risewill want to delay deliveries and extend longer credit terms to delay payment. These areknown as leads and lags and are a form of speculation.

In recent years, a number of huge losses have been incurred by speculating on themovement of exchange rates. One of the most spectacular was the case of Showaka ShellSekiyu, a Japanese oil refiner and distributor 50 percent owned by Royal Dutch Shell.From 1989 until 1992, the finance department of Showaka bet $6.44 billion worth in thefutures market that the dollar would appreciate. When the dollar depreciated (and the yenappreciated—see Figure 14.3) instead, Showaka lost $1.37 billion. More recently, therewas the five-year $750 million cumulative foreign exchange loss by John Rusnak of AllfirstBank, the U.S. subsidiary of Allied Irish Banks, Ireland’s largest bank, on trading the U.S.dollar against the Japanese yen discovered in February 2002. And in January 2004, fourforeign currency dealers at the National Australia Bank incurred losses of $360 million inthree months of unauthorized foreign exchange trades. Yes, speculation in foreign exchangeis very risky and can lead to huge losses.

14.6 Interest Arbitrage and the Efficiency of ForeignExchange Markets

Interest arbitrage refers to the international flow of short-term liquid capital to earn higherreturns abroad. Interest arbitrage can be covered or uncovered. These are discussed inturn. We will then examine the covered interest parity theory and the efficiency of foreignexchange markets.

14.6A Uncovered Interest ArbitrageSince the transfer of funds abroad to take advantage of higher interest rates in foreignmonetary centers involves the conversion of the domestic to the foreign currency to makethe investment, and the subsequent reconversion of the funds (plus the interest earned) fromthe foreign currency to the domestic currency at the time of maturity, a foreign exchangerisk is involved due to the possible depreciation of the foreign currency during the periodof the investment. If such a foreign exchange risk is covered, we have covered interestarbitrage; otherwise, we have uncovered interest arbitrage. Even though interest arbitrage isusually covered, we begin by examining the simpler uncovered interest arbitrage.

Suppose that the interest rate on three-month treasury bills is 6 percent at an annualbasis in New York and 8 percent in Frankfurt. It may then pay for a U.S. investor toexchange dollars for euros at the current spot rate and purchase EMU treasury bills to earnthe extra 2 percent interest at an annual basis. When the EMU treasury bills mature, theU.S. investor may want to exchange the euros invested plus the interest earned back intodollars. However, by that time, the euro may have depreciated so that the investor would


14.6 Interest Arbitrage and the Efficiency of Foreign Exchange Markets 445

get back fewer dollars per euro than he or she paid. If the euro depreciates by 1 percentat an annual basis during the three months of the investment, the U.S. investor nets onlyabout 1 percent from this foreign investment (the extra 2 percent interest earned minus the1 percent lost from the depreciation of the euro) at an annual basis (1/4 of 1 percent for thethree months or quarter of the investment). If the euro depreciates by 2 percent at an annualbasis during the three months, the U.S. investor gains nothing, and if the euro depreciatesby more than 2 percent, the U.S. investor loses. Of course, if the euro appreciates , the U.S.investor gains both from the extra interest earned and from the appreciation of the euro.

Related to uncovered interest arbitrage is carry trade. Carry trade refers to the strategyin which an investor borrows low-yielding currencies and lends (invest in) high-yieldingcurrencies. That is, an investor borrows a currency with a relatively low interest rate and usesthe funds to purchase another currency yielding a higher interest rate. If the higher-yieldingcurrency depreciates during the period of the investment, however, the investor runs therisk of losing money (see Case Study 14-5).

(continued)

■ CASE STUDY 14-5 Carry Trade

Earlier defined, carry trade is the strategy in whichan investor borrows a low-yielding currency andlends (invests in) a higher-yielding currency. Therisk is that if during the investment period thehigher-yielding currency depreciates vis-a-vis thelower-yielding currency by a higher percentagethan the positive interest differentials, the investorwould lose money.

For example, suppose that in a “yen carrytrade” the investor borrows yens from a Japanesebank at 1 percent interest, exchanges the yens forU.S. dollars at the prevailing dollar/yen exchangerate, and then buys a U.S. bond paying, say, 4 per-cent interest. The investor would earn 3 percent neton her or his investment—so long as the dollar/yenexchange rate does not change during the periodof the investment. If the dollar appreciated withrespect to the yen, the investor would earn thatmuch more. If, on the other hand, the dollar depre-ciated with respect to the yen during the investmentperiod, the investor would earn less, break even,or incur a loss. Specifically, if the dollar depreci-ated by less than 3 percent with respect to the yenduring the investment period, the investor wouldearn that much less. If the dollar depreciated byexactly 3 percent, the investor would break even(assuming no transaction costs). If, on the other

hand, the dollar depreciated by more than 3 per-cent, the investor’s loss would equal the differencebetween the rate of the dollar depreciation and thepositive interest differential in favor of the dollar.Thus, the big risk in carry trade is the uncertaintyof exchange rates.

Actually, the gains or losses from carry tradewill be much greater than indicated above becauseof leveraging (i.e., because the investor buys theU.S. bond on margin—puts down only a smallfraction, usually 10 percent, of the bond price). Inthis case, the gains or losses would be amplifiedtenfold.

In theory, according to uncovered interestrate parity, carry trades should not yield apredictable profit because the difference ininterest rates between two currencies shouldequal the rate at which investors expect thelow-interest-rate currency to appreciate withrespect to the high-interest-rate one. Carry trades,however, tend to weaken the currency that isborrowed, because investors sell the borrowedmoney by converting it to other currencies. Infact, the carry trade is often blamed for rapiddepreciation of the low-yielding currency andappreciation of the higher-yielding currency, thusincreasing exchange rate volatility.




The U.S. dollar and the yen have been thecurrencies most heavily used in carry trade trans-actions since the 1990s, with the yen being thelow-rate currency and U.S. dollar bonds being thehigher-yielding asset. At its peak, in early 2007,the yen carry trade was estimated to be about $1trillion. That trade largely collapsed in 2008 as aresult of the rapid appreciation of the yen. This cre-ated pressure to cover debts denominated in yen byconverting foreign assets into yen—which led to

further yen appreciation. Exchange rate volatilityoften leads to carry trade unwinds, the largest ofwhich was the 2008 one, and that contributed sub-stantially to the credit crunch that caused the 2008global financial crisis.

Source: C. Burnside, M. Eichenbaum, and S. Rebelo,“Carry Trade and Momentum in Currency Markets,”Annual Review of Financial Economics , December 2011,pp. 511–535.

14.6B Covered Interest ArbitrageInvestors of short-term funds abroad generally want to avoid the foreign exchange risk; there-fore, interest arbitrage is usually covered. To do this, the investor exchanges the domesticfor the foreign currency at the current spot rate in order to purchase the foreign treasurybills, and at the same time the investor sells forward the amount of the foreign currency heor she is investing plus the interest he or she will earn so as to coincide with the maturityof the foreign investment. Thus, covered interest arbitrage refers to the spot purchase ofthe foreign currency to make the investment and the offsetting simultaneous forward sale(swap) of the foreign currency to cover the foreign exchange risk. When the treasury billsmature, the investor can then get the domestic currency equivalent of the foreign investmentplus the interest earned without a foreign exchange risk. Since the currency with the higherinterest rate is usually at a forward discount, the net return on the investment is roughlyequal to the interest differential in favor of the foreign monetary center minus the forwarddiscount on the foreign currency. This reduction in earnings can be viewed as the cost ofinsurance against the foreign exchange risk.

As an illustration, let us continue the previous example where the interest rate onthree-month treasury bills is 6 percent per year in New York and 8 percent in Frank-furt, and assume that the euro is at a forward discount of 1 percent per year. To engagein covered interest arbitrage, the U.S. investor exchanges dollars for euros at the currentexchange rate (to purchase the EMU treasury bills) and at the same time sells forward aquantity of euros equal to the amount invested plus the interest he or she will earn at theprevailing forward rate. Since the euro is at a forward discount of 1 percent per year, theU.S. investor loses 1 percent on an annual basis on the foreign exchange transaction to coverthe foreign exchange risk. The net gain is thus the extra 2 percent interest earned minus the1 percent lost on the foreign exchange transaction, or 1 percent on an annual basis (1/4 of 1percent for the three months or quarter of the investment). Note that we express both theinterest differential and the forward discount at an annual basis and then divide by four toget the net gain for the three months or quarter of the investment.

However, as covered interest arbitrage continues, the possibility of gains diminishes untilit is completely wiped out. This occurs for two reasons. First, as funds are transferred fromNew York to Frankfurt, the interest rate rises in New York (since the supply of funds in New



York diminishes) and falls in Frankfurt (since the supply of funds in Frankfurt increases).As a result, the interest differential in favor of Frankfurt diminishes. Second, the purchaseof euros in the spot market increases the spot rate, and the sale of euros in the forwardmarket reduces the forward rate. Thus, the forward discount on the euro (i.e., the differencebetween the spot rate and the forward rate) rises. With the interest differential in favor ofFrankfurt diminishing and the forward discount on the euro rising, the net gain falls for bothreasons until it becomes zero. Then the euro is said to be at covered interest arbitrage parity(CIAP). Here, the interest differential in favor of the foreign monetary center is equal to theforward discount on the foreign currency (both expressed on an annual basis). In the realworld, a net gain of at least 1/4 of 1 percent per year is normally required to induce funds tomove internationally under covered interest arbitrage. Thus, in the preceding example, thenet annualized gain would be 3/4 of 1 percent after considering transaction costs or 0.1875percent for three months.

If the euro is instead at a forward premium, the net gain to the U.S. investor will equalthe extra interest earned plus the forward premium on the euro. However, as covered interestarbitrage continues, the interest differential in favor of Frankfurt diminishes (as indicatedearlier) and so does the forward premium on the euro until it becomes a forward discountand all of the gains are once again wiped out. Thus, the spot rate and the forward rate ona currency are closely related through covered interest arbitrage.

14.6C Covered Interest Arbitrage ParityFigure 14.4 illustrates in a more general and rigorous way the relationship, through coveredinterest arbitrage, between the interest rate differentials between two nations and the forwarddiscount or premium on the foreign currency. The vertical axis of the figure measures theinterest rate in the nation’s monetary center (i) minus the interest rate in foreign monetarycenter (i ∗), or (i−i ∗), in percentages per year. Negative values for (i –i ∗) indicate that theinterest rate is higher abroad than in the nation, while positive values indicate that theinterest rate is higher in the nation than abroad. The horizontal axis measures the forwarddiscount (–) or premium (+) on the foreign currency also expressed in percentages per year.

The solid diagonal line indicates all points of covered interest arbitrage parity (CIAP).Thus, when (i−i ∗) equals −1, the foreign currency is at a forward discount of 1 percent peryear. A positive interest differential of 1 is associated with a forward premium of 1 percent.When the interest differential is zero, the foreign currency is neither at a forward discountnor at a forward premium (i.e., the forward rate on the foreign currency is equal to its spotrate), and we are on the CIAP line at the origin.

Below the CIAP line, either the negative interest differential (in favor of the foreignmonetary center) exceeds the forward discount on the foreign currency or the forwardpremium exceeds the positive interest differential (see the figure). In either case, there willbe a net gain from a covered interest arbitrage (CIA) outflow . For example, at point A, thenegative interest differential is 2 percentage points per year in favor of the foreign monetarycenter, while the foreign currency is at a forward discount of 1 percent per year. Thus, thereis a covered interest arbitrage margin of 1 percent per year in favor of the foreign nation,leading to a capital outflow. Similarly, point A′ involves a forward premium of 2 percenton the foreign currency and a positive interest differential of only 1 percent in favor of thedomestic monetary center. Thus, investors have an incentive to invest abroad because theywould gain 2 percent on the exchange transaction and lose only 1 percent in interest in



1 2 3–1 0–2–3

1

–1

–2

–3

2

3

B

CIAInflow

CIAOutflow

CIAP

ForwardPremium

ForwardDiscount (–)

A

i – i *(%)

B'

A'

FIGURE 14.4. Covered Interest Arbitrage.The vertical axis measures the difference in the interest rate in the home nation (i) and in the foreignnation (i∗) in percentages per annum. The horizontal axis measures the forward exchange rate, with theminus sign indicating a forward discount and positive values indicating a forward premium on the foreigncurrency in percent per annum. The solid diagonal line is the covered interest parity (CIAP) line. Below theCIAP line, either the negative interest differential exceeds the forward discount or the forward premiumexceeds the positive interest differential. In either case, there will be a capital outflow under coveredinterest arbitrage. Above the CIAP line, the opposite is true and there will be an arbitrage inflow.

investing abroad. The net gain would then be 1 percent per year or 1/4 of 1 percent for thethree months or quarter of the investment.

As the arbitrage outflow continues, the net gain diminishes and tends to disappear. Specif-ically, starting from point A, the transfer of funds abroad reduces the interest differentialin favor of the foreign monetary center (say, from −2 to −1.5) and increases the forwarddiscount on the foreign currency (say, from −1 to −1.5), as explained in the previoussection, so as to reach the CIAP line (see the figure). Starting from point A′, the transfer offunds abroad will increase the positive interest differential (say, from 1 to 1.5) and reducethe forward premium (say, from +2 to +1.5) so as to once again reach the CIAP line.Specifically, as funds move abroad, interest rates tend to rise at home and decline abroad.Since interest rates were already higher at home, the positive interest differential increases.On the other hand, as investors purchase the foreign currency to invest abroad, the spot raterises. As they sell the foreign currency forward to cover their foreign exchange risk, theforward rate declines. Thus, the forward premium (i.e., the excess of the forward rate over



the spot rate) diminishes. With the positive interest differential increasing and the forwardpremium decreasing, the net gain from arbitrage outflow diminishes until it becomes zerowhen the CIAP line is reached and the arbitrage outflow comes to an end.

Above the interest parity line, either the positive interest differential exceeds the forwardpremium on the foreign currency (point B in the figure) or the negative interest differentialis smaller than the forward discount on the foreign currency (point B ′). In either case, itpays for foreigners to invest in our country, and there will be an arbitrage inflow . However,as the arbitrage inflow continues, the net gain diminishes and then disappears when theCIAP line is reached. In reality, interest arbitrage (inflow and outflow) will come to an endwhen the net gain reaches about 1/4 of 1 percent per year (1/16 of 1 percent for three months).This range is shown by the white area between the diagonal dashed lines in the figure.

14.6D Covered Interest Arbitrage MarginWe have seen that points on the CIAP line indicate either that the negative interest differential(in favor of the foreign monetary center) equals the forward discount (FD) on the foreigncurrency or that the positive interest differential (in favor of the home monetary center)equals the forward premium (FP) on the foreign currency. This can be expressed as:

i − i ∗ = FD if i < i ∗ or

i − i ∗ = FP if i > i ∗

But since the forward rate minus the spot rate divided by the spot rate [i.e., (FR−SR)/SR]measures the forward discount (if SR > FR) or the forward premium (if FR > SR), theforegoing condition for CIAP can be rewritten as

i − i ∗ = (FR − SR)/SR (14-1)

We can now define the covered interest arbitrage margin (CIAM) or the percentage gainfrom covered interest arbitrage as

CIAM = (i − i ∗) − FD or FP

or more precisely as

CIAM = (i − i ∗)/(1 + i ∗) − (FR − SR)/SR (14-2)

where (1 + i ∗) is a weighting factor. This formula is derived in the appendix to this chapter.To see how the formula works, let us apply it to the case where the interest rate on a

three-month treasury bill is 6 percent on an annual basis in New York and 8 percent inFrankfurt, while the spot rate of the euro is $1/¤1 and the three-month forward rate on theeuro is $0.99/¤1 on an annual basis. Applying the CIAM formula, we get:

CIAM = (0.06 − 0.08)/(1 + 0.08) − ($0.99 − $1.00)/$1.00

= (−0.02)/1.08 − (−$0.01)/$1.00

= −0.01852 + 0.01

= −0.00852



The negative sign for the CIAM refers to a CIA outflow or investing in Frankfurt. Theabsolute value of the CIAM indicates that the extra return per dollar invested in Frankfurtis 0.852 percent per year or 0.213 per quarter. (These values are similar to the approximatevalues obtained in the previous subsection without the weighting factor.) On a $10 millioninvestment, this means an extra return of $21,300 for investing in three-month EMU treasurybills with the foreign exchange risk covered for three months. From this extra return, wewould have to deduct the transaction costs. If these are 1/4 of 1 percent per year or 1/16 of1 percent per quarter, we get transaction costs of (0.01/16) times $10 million, which are$6,250. Thus, the net gain from investing $10 million in Frankfurt under CIA with thetransaction costs taken into consideration is $21,300 minus $6,250, or $15,050 for the threemonths of the investment.

In the real world, significant covered interest arbitrage margins are sometimes observed.The reason is not that covered interest arbitrage does not work, but it may be the result ofother forces at work. For example, higher tax rates abroad may exceed the CIAM in favorof the foreign monetary center so that no arbitrage outflows take place. Similarly, investorsmay not take advantage of a CIAM in favor of the foreign monetary center if they fear thatthe foreign government might default or impose exchange restrictions on the repatriationof profits and principal on the foreign investment. Or simply, large and persistent CIAMmay exist because of lack of information on foreign investment opportunities in developingcountries’ financial markets.

14.6E Efficiency of Foreign Exchange MarketsA market is said to be efficient if prices reflect all available information. The foreignexchange market is said to be efficient if forward rates accurately predict future spot rates;that is, if forward rates reflect all available information and quickly adjust to any newinformation so that investors cannot earn consistent and unusual profits by utilizing anyavailable information.

Questions of market efficiency are important because only when markets are efficient doprices correctly reflect the scarcity value of the various resources and result in allocationalefficiency. For example, if, for whatever reason, the price of a commodity is higher thanits value to consumers, then too many resources flow to the production of the commodityat the expense of other commodities that consumers prefer.

By their very nature, tests of market efficiency are very difficult to formulate and tointerpret. Even if foreign exchange markets were efficient, we cannot expect the forwardrate of a currency to be identical to the future spot rate of the currency because the latteralso depends on unforeseen events. However, if the forward rate exceeds the future spot rateas often as it falls below it, then we could say that the market is efficient in the sense thatthere is no available information that investors could systematically use to ensure consistentand unusual profits.

Many empirical tests have been conducted on the efficiency of foreign exchange mar-kets by Levich (1985) and others (see the references in the Selected Bibliography). Mostof these studies seem to indicate that foreign exchange markets are efficient according tothis definition of efficiency. For example, several empirical tests show that few oppor-tunities exist for risk-free arbitrage, and deviations from interest rate parity are, for themost part, smaller than transaction costs. Similarly, speculators sometimes earn profits and


14.7 Eurocurrency or Offshore Financial Markets 451

sometimes incur losses and seldom face opportunities for certain and large profits. Frankeland MacArthur (1988) have presented evidence that covered interest arbitrage holds rea-sonably well for large industrial countries but not for small ones, and Lewis (1995) showsthat the theory does not hold well for developing countries. Clarida et al. (2003) suggestthat forward-exchange-rate time series contain valuable information not yet fully utilizedfor predicting the future path of the spot exchange rates.

Thus, while many studies seem to indicate that foreign exchange markets are fairlyefficient, others do not. Exchange rates seem to respond very quickly to news, are veryvolatile, have defied all attempts at being accurately forecasted, and the variance in theirfluctuation has been relatively large, so that even if the forward rate were an unbiasedpredictor of the future spot rate it would not be an efficient one. We return to this topic inour discussion of foreign exchange forecasting in the next chapter.

As exchange rates have become more volatile, the volume of foreign exchange transac-tions has grown much faster than the volume of world trade and faster than even the muchlarger flows of investment capital. Only over the 1998–2001 period has the volume of for-eign exchange transactions declined (from $1.5 trillion in 1998 to $1.2 trillion in 2001).This was due to the introduction of the euro (which replaced several important currencies,thus eliminating the need to convert these currencies into one another) and the consolida-tion of the banking sector (which eliminated a great deal of the interbank foreign exchangemarket). More recently, the volume of foreign exchange transactions has resumed its growthand reached $4.0 billion in 2012.

14.7 Eurocurrency or Offshore Financial MarketsIn this section, we examine the operation and effects of Eurocurrency or offshore financialmarkets and also discuss Eurobonds and Euronotes.

14.7A Description and Size of the Eurocurrency MarketEurocurrency refers to commercial bank deposits outside the country of their issue. Forexample, a deposit denominated in U.S. dollars in a British commercial bank (or even ina British branch of a U.S. bank) is called a Eurodollar. Similarly, a pound sterling depositin a French commercial bank or in a French branch of a British bank is a Eurosterling, adeposit in euros (the new European currency) in a Swiss bank is simply a Eurodeposit (toavoid the awkward “Euroeuro”), and so on. These balances are usually borrowed or loanedby major international banks, international corporations, and governments when they needto acquire or invest additional funds. The market in which this borrowing and lending takesplace is called the Eurocurrency market.

Initially, only the dollar was used in this fashion, and the market was therefore calledthe Eurodollar market. Subsequently, the other leading currencies (the German mark, theJapanese yen, the British pound sterling, the French franc, and the Swiss franc) began also tobe used in this way, and so the market is now called the Eurocurrency market . The practiceof keeping bank deposits denominated in a currency other than that of the nation in whichthe deposit is held has also spread to such non-European international monetary centers asTokyo, Hong Kong, Singapore, and Kuwait, as well as to the Bahamas and the CaymanIslands in the Caribbean, and are appropriately called offshore deposits. Often, however, the



name Eurodeposits is also used for such deposits outside Europe. With these geographicalextensions, the Eurocurrency market has become an essentially 24-hour-a-day operation.Indeed, any foreign deposit made in a nation’s bank (even if in the nation’s currency) isEurocurrency if the deposit is exempted from the regulations that the nation imposes ondomestic deposits.

The Eurocurrency market consists mostly of short-term funds with maturity of less thansix months. In measuring the size of the Eurocurrency market, we must distinguish betweenits gross and net size. The first includes interbank deposits. These are the deposits of bankswith surplus Eurocurrencies to other banks facing an excess demand for Eurocurrency loans.Thus, interbank deposits represent transfers of Eurocurrency funds from one bank to anotherand do not involve a net increase in the total amount of Eurocurrency available to be lentto nonbank customers. Since the interbank market is a very important and active part of theEurocurrency market, however, the gross measure seems more appropriate in measuring thesize of the market (see Case Study 14-6).

14.7B Reasons for the Development and Growth of theEurocurrency Market

There are several reasons for the existence and spectacular growth of the Eurocurrencymarket during the past four decades. One reason was the higher interest rates often prevailingabroad on short-term deposits. Until it was abolished in March 1986, Federal ReserveSystem Regulation Q put a ceiling on the interest rates that U.S. member banks couldpay on deposits to levels that were often below the rates paid by European banks. As aresult, short-term dollar deposits were attracted to European banks and became Eurodollars.Another important reason is that international corporations often found it very convenientto hold balances abroad for short periods in the currency in which they needed to makepayments. Since the dollar is the most important international and vehicle currency in makingand receiving international payments, it is only natural for a large proportion of the currencyto be in Eurodollars. Still another reason is that international corporations can overcomedomestic credit restrictions by borrowing in the Eurocurrency market.

The Eurocurrency market originated from the desire of communist nations to keep theirdollar deposits outside the United States during the early days of the Cold War for fearthat they might be frozen in a political crisis. After 1973, the impetus to the growth of theEurodollar market came from the huge dollar deposits from petroleum-exporting countriesarising from the manyfold increases in the price of petroleum. These nations also did notwant to keep their dollar deposits in the United States for fear that the U.S. governmentmight freeze them in a political crisis. Indeed, this is exactly what happened to the (smallproportion of the) dollar deposits that Iran and Iraq did keep in the United States during theU.S. conflict with these nations in the late 1970s and early 1990s, respectively.

European banks are willing to accept deposits denominated in foreign currencies and areable to pay higher interest rates on these deposits than U.S. banks because they can lend thesedeposits at still higher rates. In general, the spread between lending and borrowing rates onEurocurrency deposits is smaller than that of U.S. banks. Thus, European banks are oftenable to pay higher deposit rates and lend at lower rates than U.S. banks. This is the result of(1) the fierce competition for deposits and loans in the Eurocurrency market, (2) the lower



■ CASE STUDY 14-6 Size and Growth of Eurocurrency Market

Table 14.4 shows the gross and the net sizeof Eurocurrency deposits (i.e., international bankdeposits denominated in currencies other than thecurrency of the borrower’s or lender’s nation) from1964 to 2011, as well as the percentage of thegross market held in the form of Eurodollars. Forcomparison purposes, the table also gives the U.S.money supply (broadly defined, or M2) over thesame period of time. The table also shows that thegross Eurocurrency market grew extremely rapidlyfrom $19 billion in 1964 to $17.9 trillion in 2007,but then it declined to $15.8 trillion as a result

■ TABLE 14.4. Size of Eurocurrency Deposit Market (in billions of dollars)

Eurodollarsas a Percentage U.S. Money Stock

Year Gross Size Net Size of Gross (M2)

1964 $19 $14 n.a. $4251968 46 34 79 5671972 210 110 78 8021976 422 199 69 1, 1521980 839 408 71 1, 5991984 1, 343 667 78 2, 3101988 2, 684 1, 083 64 2, 9941992 3, 806 1, 751 59 3, 4311996 4, 985 2, 639 55 3, 8421998 6, 332 3, 122 53 4, 4032000 6, 077 3, 532 63 4, 9492002 7, 505 4, 590 61 5, 8072004 10, 035 6, 952 57 6, 4372006 14, 168 9, 604 59 7, 0942007 17, 931 11, 966 55 7, 5222008 16, 668 10, 928 58 8, 2692009 15, 817 10, 829 57 8, 5522010 16, 014 10, 983 58 8, 8492011 16, 911 11, 374 58 9, 713

Sources: Morgan Guaranty, World Financial Markets; BIS, Quarterly Survey; and IMF, International FinancialStatistics; various issues.

of the global financial crisis, and it was $16.9 tril-lion in 2011. Gross Eurocurrency deposits thuswent from less than 5 percent of the M2 measureof the U.S. money supply in 1964 to 238 per-cent in 2007, and they were 174 percent in 2011.While the U.S. money supply grew 23 times from1964 to 2011, gross Eurocurrency deposits grew890 times! The table also shows that the percent-age of Eurodollars in gross Eurocurrency depositsdeclined from 79 percent in 1968 to 55 percent in1996 and 2007, and it was 58 percent in 2011.

operating costs in the Eurocurrency market due to the absence of legal reserve requirementsand other restrictions on Eurocurrency deposits (except for U.S. branches of Europeanbanks), (3) economies of scale in dealing with very large deposits and loans, and (4) riskdiversification. Arbitrage is so extensive in the Eurocurrency market that interest parity isgenerally maintained.



14.7C Operation and Effects of the Eurocurrency MarketOne important question is whether or not Eurocurrencies are money. Since Eurocurrenciesare, for the most part, time rather than demand deposits, they are money substitutes or nearmoney rather than money itself, according to the usual narrow definition of money or M1(which includes only currency in circulation and demand deposits). Thus, Eurobanks do not,in general, create money, but they are essentially financial intermediaries bringing togetherlenders and borrowers, and operating more like domestic savings and loan associations(before they established the so-called NOW accounts) than like commercial banks in theUnited States. Perhaps more significant is that the rapid growth of Eurocurrency depositssince the 1960s has greatly increased world liquidity. They also led to a significant integra-tion of domestic and international financial markets, which greatly increased competitionand the efficiency of domestic banking in industrialized nations.

The existence, size, and rapid growth of the Eurocurrency market have created certainproblems. One of the most serious is that the Eurocurrency market reduces the effective-ness of domestic stabilization efforts of national governments. For example, large firms thatcannot borrow domestically because of credit restrictions often can and do borrow in theEurocurrency market, thus frustrating the government effort to restrict credit to fight domes-tic inflationary pressures. This is particularly true for smaller nations where the volume ofdomestic financial transactions is small in relation to Eurocurrency transactions. A closelyrelated problem is that frequent and large flows of liquid Eurocurrency funds from oneinternational monetary center to another can produce great instability in foreign exchangerates and domestic interest rates.

Another possible problem is that Eurocurrency markets are largely uncontrolled. Asa result, a deep worldwide recession could render some of the system’s banks insolventand possibly lead internationally to the type of bank panics that afflicted capitalist nationsduring the nineteenth century, the first third of the twentieth century, and the latter partof the last decade. Domestic bank panics were more or less eliminated by the creation ofnational central banks to regulate domestic banking through deposit insurance and by settingthemselves up as “lenders of last resort” for domestic banks in a liquidity squeeze. In theEurocurrency market, however, any attempt on the part of any one nation to regulate it wouldsimply result in the market shifting its activity elsewhere. Thus, in order for regulations andguidelines to be effective, they need to be multilateral. Given the strong competition forEurobanking business, however, it is unlikely that such multilateral cooperation will beforthcoming in the near future. Indeed, nations seem to go out of their way to provide thenecessary infrastructures and eliminate existing restrictions in order to attract business.

A specific example of this is provided by the United States. Since December 1981,international banking facilities (IBFs) have been permitted in the United States. That is, U.S.banks were allowed to accept deposits from abroad and re-invest them overseas, and thuscompete directly in the huge Eurodollar market. The new rules exempted foreign deposits inU.S. banks (even if in dollars) from federally imposed reserve and insurance requirements;as such, they are Eurodollars. Several states have also passed complementary legislationexempting profits on international banking transactions from state and local taxes. Almost200 U.S. banks have entered this market, about half of them in New York, with the rest inChicago, Miami, New Orleans, and San Francisco. The United States has captured about20 percent of the huge Eurodollar market, and this has led to the creation of thousands ofnew jobs in banking, half of them in New York City.



14.7D Eurobond and Euronote MarketsEurobonds are long-term debt securities that are sold outside the borrower’s country to raiselong-term capital in a currency other than the currency of the nation where the bonds aresold. An example is provided by a U.S. corporation selling bonds in London denominated ineuros or U.S. dollars. Eurobonds have to be distinguished from foreign bonds , which refersimply to bonds sold in a foreign country but denominated in the currency of the countryin which the bonds are being sold. An example is a U.S. multinational corporation sellingbonds in England denominated in pounds sterling. Eurobonds, on the other hand, are bondssold in a foreign country and denominated in another currency. The leading centers in theinternational bond market are London, Frankfurt, New York, and Tokyo. Eurobonds differfrom most domestic bonds in that the former, as opposed to the latter, are usually unsecured(i.e., they do not require a collateral). Another type of debt security is Euronotes. Theseare medium-term financial instruments falling somewhat between short-term Eurocurrencybank loans and long-term international bonds. Corporations, banks, and countries make useof international notes to borrow medium-term funds in a currency other than the currencyof the nation in which the notes are sold.

In 2010, corporations, banks, and countries raised $1,499 billion in Eurobonds andEuronotes—down from $2,784 billion in 2007 (because of the financial crisis), but up from$200 billion in 1993. The sharp increase from 1993 to 2007 was made possible by the open-ing up of capital markets in these international debt securities by several countries, includingFrance, Germany, and Japan, and by the elimination of the U.S. interest-equalization tax.The incentive to issue Eurobonds and Euronotes is that they generally represent a lowercost of borrowing long-term funds than available alternatives. In 2010, about 72 percent ofEurobonds and Euronotes were denominated in U.S. dollars, 22 percent in euros, 2 percentin Canadian dollars, 1 percent in Australian dollars and pounds sterling, and the remaining2 percent in other currencies.

Some Eurobonds are denominated in more than one currency in order to give the lender achoice of the currencies in which to be repaid, thus providing some exchange rate protectionto the lender. For this benefit, the lender may be willing to lend at a somewhat lower rate. Alarge issue of Eurobonds and Euronotes is usually negotiated by a group (called a syndicate)of banks so as to spread the credit risk among numerous banks in many countries. Eurobondsand Euronotes usually have floating rates. That is, the interest rates charged are re-fixed,usually every three or six months, in line with changes in market conditions. After an issueof Eurobonds and Euronotes is sold by syndicate, a secondary market in the internationalnote or bond emerges in which investors can sell their holdings. (The market in which theinitial issue was sold is appropriately called the primary market .)

Interest rates on Eurocredits are expressed as a mark-up or spread over LIBOR (the Lon-don Interbank Offer Rate) or EUROBOR (the Brussels-set rate) at which Eurobanks lendfunds to one another. The spread varies according to the creditworthiness of the borrowerand ranges from 1 percent for best or prime borrowers to 2 percent for borrowers with weakcredit ratings. Often, weaker borrowers can negotiate a lower spread by paying various feesup front. These are a management fee for the bank or banks organizing the syndication, aparticipation fee to all participating banks based on the amount lent by each, as well as a com-mitment fee on any undrawn portion of the loan. Because of the size and rapid growth of theEurocurrency, and Eurobond and Euronote markets, and the resulting integration of domesticand international financial markets, we are approaching a truly global banking system.



S U M M A R Y

1. Foreign exchange markets are the markets where indi-viduals, firms, and banks buy and sell foreign cur-rencies or foreign exchange. The foreign exchangemarket for any currency, say the U.S. dollar, is com-prised of all the locations, such as London, Paris,Zurich, Frankfurt, Singapore, Hong Kong, and Tokyo,as well as New York, where dollars are bought andsold for other currencies. These different monetary cen-ters are connected by a telephone network and videoscreens, and are in constant contact with one another.

2. The principal function of foreign exchange markets isthe transfer of purchasing power from one nation andcurrency to another. The demand for foreign exchangearises from the desire to import or purchase goodsand services from other nations and to make invest-ments abroad. The supply of foreign exchange comesfrom exporting or selling goods and services to othernations and from the inflow of foreign investments.About 90 percent of all foreign exchange transac-tions today, however, are undertaken by foreign cur-rency traders and speculators. A nation’s commer-cial banks operate as clearinghouses for the foreignexchange demanded and supplied. Commercial banksthen even out their excess supply of or demand for for-eign exchange with other commercial banks throughthe intermediation of foreign exchange brokers. Thenation’s central bank then acts as the lender or bor-rower of last resort.

3. The exchange rate (R) is defined as the domestic cur-rency price of the foreign currency. Under a flexibleexchange rate system of the type in existence since1973, the equilibrium exchange rate is determinedat the intersection of the nation’s aggregate demandand supply curves for the foreign currency. If thedomestic currency price of the foreign currency rises,we say that the domestic currency depreciated. Inthe opposite case, we say that the domestic currencyappreciated (and the foreign currency depreciated).Arbitrage refers to the purchase of a currency whereit is cheaper for immediate resale where it is moreexpensive in order to make a profit. This equalizesexchange rates and ensures consistent cross ratesin all monetary centers, unifying them into a sin-gle market. Under a managed floating exchange ratesystem (of the type in operation today), the loss ofofficial reserves only indicates the degree of official

intervention in foreign exchange markets to influencethe level and movement of exchange rates, and notthe balance-of-payments deficit.

4. A spot transaction involves the exchange of curren-cies for delivery within two business days. A forwardtransaction is an agreement to purchase for deliveryat a future date (usually one, three, or six monthshence) a specified amount of a foreign currency ata rate agreed upon today (the forward rate). Whenthe forward rate is lower than the spot rate, the for-eign currency is said to be at a forward discount of acertain percentage per year. In the opposite case, theforeign currency is said to be at a forward premium.Currency swap is a spot sale of a currency combinedwith a forward repurchase of the same currency. Aforeign exchange futures is a forward contract forstandardized currency amounts and selected calendardates traded on an organized market (exchange). Aforeign exchange option is a contract specifying theright to buy or sell a standard amount of a tradedcurrency at or before a stated date.

5. Because exchange rates usually change over time,they impose a foreign exchange risk on anyone whoexpects to make or receive a payment in a foreigncurrency at a future date. The covering of such anexchange risk is called hedging. Speculation is theopposite of hedging. It refers to the taking of an openposition in the expectation of making a profit. Specu-lation can be stabilizing or destabilizing. Hedging andspeculation can take place in the spot, forward, future,or options markets—usually in the forward market.

6. Interest arbitrage refers to the international flow ofshort-term liquid funds to earn higher returns abroad.One aspect of this is carry trade. Covered interestarbitrage refers to the spot purchase of the foreigncurrency to make the investment and an offsettingsimultaneous forward sale of the foreign currency tocover the foreign exchange risk. The net return fromcovered interest arbitrage is usually equal to the inter-est differential in favor of the foreign monetary centerminus the forward discount on the foreign currency.As covered interest arbitrage continues, the net gainis reduced and finally eliminated. When the net gainis zero, the currency is said to be at interest parity.Foreign exchange markets are said to be efficient ifforward rates accurately predict future spot rates.


Questions for Review 457

7. Eurocurrency refers to commercial bank depositsoutside the country of their issue, or even in thesame country, if exempted from regulations imposedon domestic deposits. The Eurocurrency market hasgrown very rapidly during the past three decades. Thereasons for its existence and growth are (1) the higherinterest rates paid on Eurocurrency deposits, (2) theconvenience it provides for international corporations,and (3) the ability to escape national monetary con-trols. Eurobanks do not, in general, create money,

but they are essentially financial intermediaries bring-ing together lenders and borrowers. The Eurocurrencymarket can create great instability in exchange andother financial markets. Eurobonds are long-term debtsecurities sold outside the borrower’s country in acurrency other than the currency of the nation wherethe bonds are sold. Euronotes are medium-term finan-cial instruments falling somewhat between short-termEurocurrency and Eurobonds.

A L O O K A H E A D

The next chapter examines exchange rate determination,both in the long run and in the short run, and discussesthe reasons for the large exchange rate disequilibria and

great volatility in foreign exchange markets during the pastthree decades. We also evaluate the accuracy of exchangerate forecasting in the real world.

K E Y T E R M S

Appreciation, p. 429Arbitrage, p. 431Carry trade, p. 445Covered interest

arbitrage,p. 444

Covered interestarbitrage margin(CIAM), p. 449

Covered interestarbitrage parity(CIAP),p. 447

Cross-exchange rate,p. 429

Depreciation, p. 429Destabilizing

speculation,p. 443

Effective exchangerate, p. 431

Efficiency of foreignexchangemarkets, p. 450

Euro, p. 427Eurobonds, p. 455

Eurocurrency,p. 451

Eurocurrencymarket, p. 451

Euronotes, p. 455Exchange rate,

p. 428Foreign exchange

futures, p. 436Foreign exchange

market, p. 423Foreign exchange

option, p. 436

Foreign exchangerisk, p. 440

Foreign exchangeswaps, p. 435

Forward discount,p. 435

Forward premium,p. 435

Forward rate,p. 434

Hedging, p. 441Interest arbitrage,

p. 444

Offshore deposits,p. 451

Seignorage, p. 425Speculation, p. 442Spot rate, p. 434Stabilizing

speculation,p. 443

Uncovered interestarbitrage,p. 444

Vehicle currency,p. 425

Q U E S T I O N S F O R R E V I E W

1. What are foreign exchange markets? What is theirmost important function? How is this function per-formed?

2. What are the four different levels of participantsin foreign exchange markets? What are the otherfunctions of foreign exchange markets?

3. What is meant by the exchange rate? How is theequilibrium exchange rate determined under a flex-ible exchange rate system?

4. What is meant by a depreciation of the domestic cur-rency? an appreciation? What is the cross exchangerate? What is the effective exchange rate?

5. What is arbitrage? What is its result? What is tri-angular arbitrage? What are cross rates?

6. Why is the measure of the balance-of-paymentsdeficit or surplus not strictly appropriate under aflexible exchange rate?

7. What is meant by a spot transaction and the spotrate? a forward transaction and the forward rate?What is meant by a forward discount? forward pre-mium? What is a currency swap? What is a foreignexchange futures? a foreign exchange option?



8. What is meant by foreign exchange risk? Howcan foreign exchange risks be covered in the spot,forward, futures, or options markets? Why doeshedging not usually take place in the spotmarket?

9. What is meant by speculation? How can specu-lation take place in the spot, forward, futures, oroptions markets? Why does speculation not usuallytake place in the spot market? What is stabilizingspeculation? destabilizing speculation?

10. What is interest arbitrage? uncovered interest arbi-trage? covered interest arbitrage? How is inter-est arbitrage covered in the forward market? Whydoes the net gain from covered interest arbi-trage tend to diminish as covered interest arbitragecontinues?

11. What is meant by the foreign currency being atcovered interest arbitrage parity (CIAP)? What aresome of the forces that can prevent the achievementof CIAP?

12. (a) What is a Eurocurrency? (b) Why would off-shore deposits be a more appropriate term? (c) Whyis the spread between lending and borrowing rateslower on Eurocurrencies than on commercial bankdollar deposits in the United States?

13. (a) Are Eurocurrencies money? (b) Do Eurobankscreate money? (c) What are the most serious prob-lems created by the existence of Eurocurrencies?

14. What is the difference between Eurocurrencies onthe one hand and Eurobonds and Euronotes on theother?

P R O B L E M S

1. From the following figure, determine

(a) the equilibrium exchange rate between thedollar and the pound sterling and the equilibriumquantity of pounds with supply curve S£ and S ′

£under a flexible exchange rate system.

(b) If the United States wanted to maintain theexchange rate at $3 = £1 with supply curve S£,how much pound reserves would the U.S. centralbank gain or lose per day?

2. sfasfd(a) Redraw demand curve for pounds D£ andsupply curve of pounds S£ as in the figure ofProblem 1 and draw on it another supply curvefor pounds (label it S ∗

£) that intersects D£ at $1= £1 (label the point of intersection C).

B

A

S'

D

3

2

1

0 20 40 60 Million /Day

R = $/

S

(b) Assuming a flexible exchange rate system,determine the equilibrium exchange rate and equi-librium quantity of pounds with S ∗

£.

(c) If the United States wanted to maintain afixed exchange rate of R = 1.5 with S ∗

£, indicatethe amount of pound reserves that the U.S. centralbank would gain or lose per day.

3. Assume the following exchange rates:

$2 = £1 in New York

¥410 = £1 in London

¥200 = $1 in Tokyo

Indicate how profitable triangular, or three-point,arbitrage can take place.

4. sfasfd(a) Identify the forces at work that will make thecross exchange rates consistent in currency arbi-trage in the previous problem.

(b) What are the consistent cross-rates in Prob-lem 3?

*5. Calculate the forward discount or premium for thefollowing spot and three-month forward rates:

(a) SR = $2.00/£1 and FR = $2.01/£1

(b) SR = $2.00/£1 and FR = $1.96/£1


A14.1 Derivation of the Formula for the Covered Interest Arbitrage Margin 459

6. Calculate the forward discount or premium for thefollowing spot and three-month forward rates:

(a) SR = SF2/¤1 and SF2.02/¤1 where SF isthe Swiss franc and ¤ is the euro

(b) SR = ¥200/$1 and FR = ¥190/$1

7. Assume that SR = $2/£1 and the three-month FR= $1.96/£1. How can an importer who will haveto pay £10,000 in three months hedge the foreignexchange risk?

8. For the given in Problem 7, indicate how anexporter who expects to receive a payment of£1 million in three months hedges the foreignexchange risk.

*9. Assume that the three month FR = $2.00/£1 anda speculator believes that the spot rate in threemonths will be SR = $2.05/£1. How can a personspeculate in the forward market? How much willthe speculator earn if he or she is correct?

10. If the speculator of Problem 9 believes that thespot rate in three months will be SR = $1.95/£1,how can he or she speculate in the forward mar-ket? How much will the speculator earn if he orshe is correct? What will the result be if in threemonths SR = $2.05/£1 instead?

*11. If the positive interest rate differential in favor ofa foreign monetary center is 4 percent per yearand the foreign currency is at a forward discountof 2 percent per year, roughly how much would aninterest arbitrageur earn from the purchase of for-eign three-month treasury bills if he or she coveredthe foreign exchange risk?

12. For the given of Problem 11, indicate:

(a) How much would an interest arbitrageur earnif the foreign currency were at a forward premiumof 1 percent per year?

(b) What would happen if the foreign currencywere at a forward discount of 6 percent per year?

13. With reference to Figure 14.4, explain (a) whythere will be an arbitrage inflow at points B andB ′ and (b) the forces that tend to eliminate the netgain as arbitrage inflows continue.

14. Explain why even when CIAP holds, investorsin different monetary centers do not necessarilyreceive the same returns on their financial invest-ments.

*= Answer provided at www.wiley.com/college/salvatore.

APPENDIX

A14.1 Derivation of the Formula for the Covered InterestArbitrage Margin

In this appendix, we derive the formula for calculating the covered interest arbitrage margin(CIAM). Starting with Formula (14A-1):

K (1 + i/4)>=<

(K /SR)(1 + i ∗/4)FR (14A-1)

where K = amount of capital investedi = domestic interest rate per year

i ∗ = foreign interest rate per yearSR = spot rateFR = forward rate

The left-hand side of Formula (14A-1) gives the value of the investment (the originalcapital invested plus the interest earned) when K amount of capital is invested domesticallyfor three months. The right-hand side gives the domestic currency value of the investment(the original capital invested plus the interest earned) when the same amount of capital is

http://www.wiley.com/college



invested abroad for three months with the foreign exchange risk covered. Specifically, theright-hand side of the formula gives the foreign currency value of the investment, timesone plus the interest earned abroad for three months, times the forward rate (to reconvertthe invested capital plus the interest earned back into the domestic currency). Investors willinvest domestically if the left-hand side of the formula is larger than the right-hand side;they will invest abroad if the right-hand side of the formula is larger than the left-hand side;and they are indifferent if the two sides are equal.

According to the theory of covered interest arbitrage (CIA), an arbitrage outflow orinflow will proceed until no net gain remains (i.e., until covered interest arbitrage parity orCIAP is reached). Thus, at CIAP the two sides of Formula (14A-1) would be equal. Bytreating Formula (14A-1) as an equation and manipulating it algebraically, we can derivethe formula for the covered interest arbitrage margin (CIAM). In doing this, it is convenientto divide both sides by K and omit, for the moment, the division of i and i ∗ by 4. Thiswill give us the CIAM per dollar per year. By then multiplying the CIAM obtained by thecapital invested (K ) and dividing by 4, we get the extra dollar earnings in percentage forthe three months of the investment with the foreign exchange risk covered.

Treating Formula (14A-1) as an equation, and dividing both sides by K and omitting thedivision of i and i ∗ by 4 as explained previously, we get Equation (14A-2):

1 + i = (FR/SR) (1 + i ∗) (14A-2)

Manipulating Equation (14A-2) algebraically, we obtain:

(1 + i )/(1 + i ∗) = FR/SR

[(1 + i )/(1 + i ∗)] − 1 = [FR/SR] − 1

(1 + i − 1 − i ∗)/(1 + i ∗) = (FR − SR)/SR

(i − i ∗)/(1 + i ∗) = (FR − SR)/SR

Solving for FR, we get Formula (14A-3) to calculate the forward rate at CIAP:

FR = [(i − i ∗)/(1 + i ∗)]SR + SR (14A-3)

Thus, the covered interest arbitrage margin (CIAM) is:

CIAM = (i − i ∗)/(1 + i ∗) − (FR − SR)/SR (14A-4)

This is Formula (14-2) given in Section 14.6d. The first fraction on the right-hand sideof Formula (14A-4) is the domestic-foreign interest rate differential weighted by 1 plus theforeign interest rate. The second fraction is the forward discount on the foreign currencyweighted by the spot rate. At CIAP, the value of the two fractions is equal, so that CIAMequals zero. Since Formula (14A-4) refers to a whole year, the CIAM for three months isCIAM/4.

Problem Using the data on i , i ∗, SR, and FR in the numerical example at the end ofSection 14.6d, determine the dollar amount (principal plus interest) that a U.S. investorwill get back by investing $100,000 for three months in (a) U.S. treasury bills or (b) EMUtreasury bills with the foreign exchange risk covered. How would your answer to Part (b)differ if you estimated the CIAM as in Section 14.6b?


Selected Bibliography 461

S E L E C T E D B I B L I O G R A P H Y

The operation of the foreign exchange markets is explained indetail in:

■ S. Y. Cross, All about the Foreign Exchange Market in the

United States (New York: Federal Reserve Bank of New York,

1998).

■ The Federal Reserve Bank of New York, Survey of Foreign

Exchange Market (New York: Federal Reserve Bank of New

York, 1998).

■ R. Z. Aliber, The International Money Game, 6th ed. (New

York: Basic Books, 2000).

■ Federal Reserve Bank of Chicago, Readings in International

Finance, 3rd ed. (Chicago: Federal Reserve Bank of Chicago,

1987), Part VI.

■ R. M. Levich, International Financial Markets: Prices and

Policies (Boston: Irwin McGraw-Hill, 1998).

■ P. Hartman, Currency Competition and Foreign Exchange

Markets (Cambridge: Cambridge University Press, 1999).

■ International Monetary Fund, International Capital Markets

(Washington, D.C.: IMF, 2012).

■ Bank for International Settlements, Annual Report (Basel,

Switzerland: BIS, 2012).

On the creation of the euro and the international use of the dollar,the euro, and the yen, see:

■ G. Tavlas and Y. Ozeki, The Internationalization of Curren-

cies: An Appraisal of the Japanese Yen , Occasional Paper

90 (Washington, D.C.: International Monetary Fund, January

1992).

■ G. Tavlas, “The International Use of Currencies: The U.S.

Dollar and the Euro,” Finance and Development , June 1998,

pp. 46–49.

■ Board of Governors of the Federal Reserve System, “The

Launch of the Euro,” Federal Reserve Bulletin , October 1999,

pp. 655–666.

■ D. Salvatore, “The Euro: Expectations and Performance,”

Eastern Economic Journal , Winter 2002, pp. 121–136.

■ P. Bacchetta and E. van Wincoop, “A Theory of the Curency

Denomination of International Trade,” NBER Working Paper

No. 9039 , July 2002.

■ M. D. Chinn and J. A. Frankel, “The Euro May Surpass the

Dollar as the Leading Currency over the Next 15 Years,”

NBER Paper No. 13909 , April 2008.

■ L. S. Goldberg and C. Tille, “Vehicle Currency Use in Interna-

tional Trade,” Journal of International Economics , December

2008, pp. 177–192.

■ D. Salvatore, “Euro” and “Vehicle Currency,” Princeton Ency-

clopedia of the World Economy (Princeton, N.J.: Princeton

University Press, 2008), pp. 350–352, 1162–1163.

On the internationalization of the renminbi or yuan, see:

■ D. Salvatore, “Evolution of Chinese Financial Mar-

kets in the Global Context,” The Chinese Economy .

November–December 2010, pp. 8–21.

■ B. Eichengreen, “The Renminbi as an International Cur-

rency,” Journal of Policy Modeling , September/October 2011,

pp. 723–730.

■ IMF, Internationalization of Emerging Market Currencies: A

Balance between Risks and Rewards (Washington, D.C.: IMF

Staff Discussion Note, October 19, 2011).

■ R. McCauley, “Renminbi Internationalization and China’s

Financial Development,” BIS Quarterly Review , December

2011, pp. 41–56.

■ D. Salvatore, “Exchange Rate Misalignments and the Inter-

national Monetary System,” Journal of Policy Modeling ,

July/august 2012, pp. 594–604.

For the carry trade, see:

■ C. Burnside, “Carry Trade and Risks,” NBER Working Paper

No. 17278 , August 2011.

■ C. Burnside, M. Eichenbaum, and S. Rebelo, “Carry Trade

and Momentum in Currency Markets,” Annual Review of

Financial Economics , December 2011, pp. 511–535.

The classic work on the spot and forward markets for foreignexchange, hedging, and speculation is:

■ P. Einzig, The Dynamic Theory of Forward Exchange (Lon-

don: Macmillan, 1967).

The first clear exposition of the theory of covered interest arbi-trage is found in:

■ J. M. Keynes, A Tract on Monetary Reform (London: Macmil-

lan, 1923), pp. 113–139.

For the empirical testing of covered interest arbitrage and theefficiency of foreign exchange markets, see:

■ J. A. Frankel and M. L. Mussa, “The Efficiency of the Foreign

Exchange Market and Measures of Turbulence,” American

Economic Review , May 1980, pp. 374–381.

■ J. F. O. Bilson, “The ‘Speculative Efficiency’ Hypothesis,”

Journal of Business , July 1981, pp. 435–451.

■ R. M. Levich, “Empirical Studies of Exchange Rates: Price

Behavior, Rate Determination and Market Efficiency”, in



R. W. Jones and P. B. Kenen, eds., Handbook of Interna-tional Economics , Vol. II (Amsterdam: North-Holland, 1985),pp. 979–1040.

■ J. A. Frankel and A. T. MacArthur, “Political vs. CurrencyPremia in International Real Interest Differentials,” EuropeanEconomic Review , June 1988, pp. 1083–1121.

■ D. L. Thornton, “Tests of Covered Interest Arbitrage,” Fed-eral Reserve Bank of St. Louis , July–August 1989, pp. 55–66.

■ K. A. Lewis, “Puzzles in International Financial Markets,” inG. M. Grossman and K. Rogoff, eds., Handbook of Inter-national Economics , Vol. III (Amsterdam: North-Holland,1995), pp. 1913–1971.

■ M. Taylor, “The Economics of Exchange Rates,” Journal ofEconomic Literature, March 1995, pp. 13–47.

■ K. Rogoff, “Monetary Models of the Dollar/Yen/Euro,” TheEconomic Journal , November 1999, pp. 655–659.

■ R. H. Clarida, L. Sarno, M. P. Taylor, and G. Valente, “TheOut of Sample Success of Term Structure Models as ExchangeRate Predictors: A Step Beyond,” Journal of InternationalEconomics , Vol. 60, 2003, pp. 61–83.

■ M. Chinn, “The (Partial) Rehabilitation of Interest Rate Par-ity in the Floating Rate Era: Longer Horizons, AlternativeExpectations, and Emerging Markets,” Journal of Interna-tional Money and Finance, February 2006, pp. 7–21.

A discussion of the operation of the Euromarkets and interna-tional banking is found in:

■ G. McKenzie, Economics of the Eurodollar Market (London:

Macmillan, 1976).

■ R. I. McKinnon, The Eurocurrency Market , Princeton Essays

in International Finance, No. 125 (Princeton, N.J.: Prince-

ton University Press, International Finance Section, December

1977).

■ Y. S. Park and J. Zwick, International Banking in Theory and

Practice (Reading, Mass.: Addison-Wesley, 1985).

■ Federal Reserve Bank of Chicago, Reading in International

Finance, 3rd ed. (Chicago: Federal Reserve Bank of Chicago,

1987), Parts IV and V.

■ R. L. Heinkel and M. D. Levi, “The Structure of International

Banking,” Journal of International Money and Finance, June

1992, pp. 251–272.

■ International Monetary Fund, Modern Banking and OTC

Derivatives Markets (Washington, D.C.: IMF, 2000).

■ M. Obstfeld and A. M. Taylor, “Globalization and Capital

Markets,” NBER Working Paper No. 8846 , March 2002.

■ Bank of International Settlements, Triennial Central Bank Sur-

vey (Basel, Switzerland: BIS, March 2012).

■ International Monetary Fund, International Capital Markets

(Washington, D.C.: IMF, 2012).

■ Bank for International Settlements, Annual Report (Basel,

Switzerland: BIS, 2012).

I N T E R N e t

The basics of the foreign exchange market in the UnitedStates can be found at:

http://www.newyorkfed.org/education/addpub/usfxm/chap2.pdf

Data on exchange rates by country and region, cross-rates,and the ability to calculate the exchange rate between anytwo currencies can be found at:

http://www.x-rates.com/

The monthly trade-weighted exchange rate of the dollar,as well as data on U.S. interest rates, can be obtained byclicking, respectively, on “Exchange Rates” and “Inter-est Rates” (for covered interest arbitrage) on the FederalReserve Bank of St. Louis website at:

http://research.stlouisfed.org/fred2

For the internationalization of the renminbi, see:

http://www.bis.org/repofficepubl/arpresearch200903.05.pdf

http://www.citibank.com/transactionservices/home/corporations/docs/rmb.pdf

http://www.jpmorgan.com/tss/General/China_Internationalization_of_RMB/1288220029583

The carry trade is examined at:

http://www.babypips.com/school/what-is-carry-trade.html

http://www.fxwords.com/c/carry-trade.htm

http://www.moneyweek.com/investments/why-is-the-carry-trade-so-dangerous

http://www.newyorkfed.org/education/addpub/usfxm

http://www.x-rates.com

http://research.stlouisfed.org/fred2

http://www.bis.org/repofficepubl/arpresearch200903

http://www.citibank.com/transactionservices/home

http://www.jpmorgan.com/tss/General/China_

http://www.babypips.com/school/what-is-carry-trade

http://www.fxwords.com/c/carry-trade.htm




Salvatore c15.tex V2 - 10/18/2012 12:45 A.M. Page 463

Exchange Rate Determination chapter

LEARNING GOALS:

After reading this chapter, you should be able to:

• Understand the purchasing-power parity theory andwhy it does not work in the short run

• Understand how the monetary and the portfolio balancemodels of the exchange rate work

• Understand the causes of exchange rate overshooting

• Understand why exchange rates are so difficult toforecast

15.1 IntroductionIn this chapter, we examine modern exchange rate theories. These theories arebased on the monetary approach and the asset market or portfolio balance approachto the balance of payments that have been developed since the late 1960s. Thesetheories view the exchange rate, for the most part, as a purely financial phenomenon.They also seek to explain the great short-run volatility of exchange rates and theirtendency to overshoot their long-run equilibrium level, which have often beenobserved during the past four decades.

These modern exchange rate theories may be distinguished from traditionalexchange rate theories (discussed in Chapters 16 and 17), which are based on tradeflows and help explain exchange rate movements only in the long run or over theyears. Since the advent of floating rates in 1973, international financial flows haveincreased tremendously and are now far larger than trade flows. Therefore, it isonly natural that interest shifted toward monetary theories of exchange rate deter-mination. Traditional exchange rate theories are still important, however, especiallyin explaining exchange rates in the long run.

We begin in Section 15.2 by presenting the purchasing-power parity theory,which provides the long-run framework for the monetary and asset market orportfolio balance approaches to exchange rate determination. Section 15.3 then

463


464 Exchange Rate Determination

examines the monetary approach to the balance of payments and exchange rate determina-tion. Section 15.4 presents the portfolio balance approach to exchange rate determination.Section 15.5 examines exchange rate dynamics and seeks to explain the tendency of short-runexchange rates to overshoot their long-run equilibrium level. Finally, Section 15.6 presentsempirical evidence on the monetary approach and the portfolio balance approach, and onexchange rate forecasting. The appendix to the chapter discusses a formal model of themonetary approach and portfolio balance model of exchange rate determination.

15.2 Purchasing-Power Parity TheoryIn this section, we examine the purchasing-power parity (PPP) theory and evaluate itsusefulness in explaining exchange rates. The purchasing-power parity (PPP) theory waselaborated and brought back into use by the Swedish economist Gustav Cassel in order toestimate the equilibrium exchange rates at which nations could return to the gold standardafter the disruption of international trade and the large changes in relative commodity pricesin the various nations caused by World War I. There is an absolute and a relative versionof the PPP theory. These will be examined in turn.

15.2A Absolute Purchasing-Power Parity TheoryThe absolute purchasing-power parity theory postulates that the equilibrium exchange ratebetween two currencies is equal to the ratio of the price levels in the two nations. Specifi-cally:

R = P

P∗ (15-1)

where R is the exchange rate or spot rate and P and P∗ are, respectively, the general pricelevel in the home nation and in the foreign nation. For example, if the price of one bushel ofwheat is $1 in the United States and ¤1 in the European Monetary Union, then the exchangerate between the dollar and the pound should be R = $1/¤1 = 1. That is, according to thelaw of one price, a given commodity should have the same price (so that the purchasingpower of the two currencies is at parity) in both countries when expressed in terms of thesame currency. If the price of one bushel of wheat in terms of dollars were $0.50 in theUnited States and $1.50 in the European Monetary Union, firms would purchase wheat inthe United States and resell it in the European Monetary Union, at a profit. This commodityarbitrage would cause the price of wheat to fall in the European Monetary Union and rise inthe United States until the prices were equal, say $1 per bushel, in both economies (in theabsence of obstructions to the flow of trade or subsidies and abstracting from transportationcosts). Commodity arbitrage thus operates just as does currency arbitrage in equalizingcommodity prices throughout the market.

This version of the PPP theory can be very misleading. There are several reasons for this.First, it appears to give the exchange rate that equilibrates trade in goods and services whilecompletely disregarding the capital account. Thus, a nation experiencing capital outflowswould have a deficit in its balance of payments, while a nation receiving capital inflowswould have a surplus if the exchange rate were the one that equilibrated international trade ingoods and services. Second, this version of the PPP theory will not even give the exchange


15.2 Purchasing-Power Parity Theory 465

rate that equilibrates trade in goods and services because of the existence of many nontradedgoods and services.

Nontraded goods include products, such as cement and bricks, for which the cost oftransportation is too high for them to enter international trade, except perhaps in borderareas. Most services, including those of mechanics, hair stylists, family doctors, and manyothers, also do not enter international trade. International trade tends to equalize the pricesof traded goods and services among nations but not the prices of nontraded goods andservices. Since the general price level in each nation includes both traded and nontradedcommodities, and prices of the latter are not equalized by international trade, the absolutePPP theory will not lead to the exchange rate that equilibrates trade. Furthermore, theabsolute PPP theory fails to take into account transportation costs or other obstructions tothe free flow of international trade. As a result, the absolute PPP theory cannot be taken tooseriously (see Case Studies 15-1 and 15-2). Whenever the purchasing-power parity theoryis used, it is usually in its relative formulation.

15.2B Relative Purchasing-Power Parity TheoryThe more refined relative purchasing-power parity theory postulates that the change in theexchange rate over a period of time should be proportional to the relative change in theprice levels in the two nations over the same time period. Specifically, if we let the subscript0 refer to the base period and the subscript 1 to a subsequent period, the relative PPP theorypostulates that

R1 = P1/P0

P∗1/P∗

0· R0 (15-2)

where R1 and R0 are, respectively, the exchange rates in period 1 and in the base period.For example, if the general price level does not change in the foreign nation from the

base period to period 1 (i.e., P∗1/P∗

0 = 1), while the general price level in the home nationincreases by 50 percent, the relative PPP theory postulates that the exchange rate (definedas the home-currency price of a unit of the foreign nation’s currency) should be 50 percenthigher (i.e., the home nation’s currency should depreciate by 50 percent) in period 1 ascompared with the base period.

Note that if the absolute PPP held, the relative PPP would also hold, but when the relativePPP holds, the absolute PPP need not hold. For example, while the very existence of capitalflows, transportation costs, other obstructions to the free flow of international trade, andgovernment intervention policies leads to the rejection of the absolute PPP, only a changein these would lead the relative PPP theory astray.

However, other difficulties remain with the relative PPP theory. One of these resultsfrom the fact (pointed out by Balassa and Samuelson in 1964) that the ratio of the price ofnontraded to the price of traded goods and services is systematically higher in developednations than in developing nations. The Balassa–Samuelson effect results from laborproductivity in traded goods being higher in developed than in developing countries, butabout the same in many nontraded goods and services sectors (for example, haircutting). Toremain in nontraded goods and services sectors in developed nations, however, labor mustreceive wages comparable to the high wages in traded -goods sectors. This makes the priceof nontraded goods and services systematically higher in developed than in developing



■ CASE STUDY 15-1 Absolute Purchasing-Power Parity in the Real World

Figure 15.1 shows the actual exchange rate of thedollar in terms of the German mark (i.e., DM/$prevailing in the market—the colored curve) andthe PPP exchange rate (measured by the ratio ofthe German to the U.S. consumer price index—theblack curve) during the flexible exchange rateperiod since 1973. (Since the beginning of 1999,the fluctuation of the DM/$ reflects the fluctua-tion of the euro with respect to the dollar.) Forthe absolute PPP theory to hold, the two curvesshould coincide. As we can see from the figure,

1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993

1.4

1.6

1.8

2.0

2.2

2.4

PPP

Actual

Overvalued dollar

Undervalued dollar

2.6

2.8

3.0

1.2

3.2

1995 1997 1999 2001 2003 2005

Year

DM/$

2007 2009 2011

FIGURE 15.1. Actual and PPP Exchange Rate of the Dollar, 1973–2011.The colored curve measures the dollar exchange rate (defined as DM/$) prevailing in the market, and the black curvemeasures the PPP exchange rate (measured by the ratio of the German to the U.S. consumer price index) from 1973 to2011. The figure shows that the dollar was undervalued during 1973–1980, 1986–2000, and 2003–2011, and was overvaluedduring 1981–1985 and in 2001 and 2002. (Since the beginning of 1999, the fluctuation in DM/$ reflects the fluctuation of theeuro with respect to the dollar.)Source: International Monetary Fund, International Financial Statistics (Washington, D.C.: IMF, various issues).

however, the curves diverge widely. The dollarwas undervalued (the colored curve was below theblack curve) from 1973 to 1980, 1986 to 2000,and 2003 to 2011, and was overvalued from 1981to 1985 and 2001 and 2002. The figure shows thatat its peak (at the beginning of 1985), the dollarwas overvalued by nearly 40 percent in terms ofmarks. Only at the beginning of 1981 and 2001,and at the end of 1985 and 2002, do the curvescross and the two currencies were at parity.

nations. For example, the price of a haircut may be $10 in the United States but only $1 inBrazil.

Since the general price index includes the prices of both traded and nontraded goods andservices, and prices of the latter are not equalized by international trade but are relativelyhigher in developed nations, the relative PPP theory will tend to predict overvalued exchangerates for developed nations and undervalued exchange rates for developing nations, withdistortions being larger the greater the differences in the levels of development. This hasbeen confirmed by Rogoff (1996) and Choudri and Khan (2005).



■ CASE STUDY 15-2 The Big Mac Index and the Law of One Price

According to the absolute PPP theory, the dollarprice of a particular product—say, McDonald’s BigMac hamburger—should be the same in other coun-tries as in the United States if exchange rates wereequal to the ratio of the price level in the UnitedStates and other countries. From the second columnin Table 15.1, however, we see that the dollar priceof a Big Mac varied greatly across countries. OnJanuary 12, 2012, the Big Mac was most expensivein Norway ($6.79) and cheapest in India ($1.62), ascompared with $4.20 in the United States.

The third column of the table gives the im-plied purchasing-power parity (PPP) of the dollarwith respect to the various currencies. This is theexchange rate that would make the price of a ham-burger the same in the various countries or regionsas in the United States. For example, the price of£3.49 for a hamburger in the Euro area implies adollar-euro exchange rate of 1.2034 (rounded offto 1.20 in Table 15.1) to equalize the price of ahamburger of $4.20 (£3.49 × 1.2034 = $4.20)

■ TABLE 15.1. Big Mac Prices and Exchange Rates, January 12, 2012

Big Mac PricesActual Dollar Under (−)/Over (+)

In Local In U.S. Implied PPPa Exchange Rate: against theCurrency Dollars of the Dollar Jan. 12, 2012 Dollar, %

United Statesb $4.20 $4.20 − − −Argentina Peso 20.0 $4.64 4.77 4.31 10Australia A$4.80 $4.54 1.14 0.97 18Brazil Real 10.25 $5.68 2.44 1.81 35Britainc £2.49 $3.82 1.65 1.54 −5Canada C$4.73 $4.63 113 1.02 10Chile Peso 2,050 $4.05 488 505 −3Chinad Yuan 15.4 $2.44 3.67 6.32 −42Colombia Preso 8,400 $4.54 2001 1852 8Czech Republic Koruna 70.22 $3.45 15.73 20.4 −18Denmark DK 31.5 $5.37 7.50 5.86 28Egypt Pound 15.5 $2.57 3.69 6.04 −39Euro areae 3.49 $4.43 1.20 1.27f 5Hong Kong HK$16.5 $2.12 3.93 7.77 −45Hungary Forint 64.5 $2.63 153.67 245 −37India Rupee 84.0 $1.62 20.01 51.9 −61

(continued)

in the two regions. This makes the actual dollar-euro exchange rate of $1.27/£ about 6 percent[(1.27 − 1.2034)/1.2034 = 5.53 percent, roundedoff to 6 percent in Table 15.1] overvalued withrespect to the dollar.

Since the dollar price of a Big Mac was$6.79 in Norway as compared with $4.20 in theUnited States, the Norwegian kroner was 62 per-cent ($6.79/$4.20) overvalued with respect to theU.S. dollar on January 12, 2012. The table alsoshows that the Swiss franc was also overvaluedby 62 percent, the Swedish krone by 41 percent,and the Brazilian real by 35 percent. On the otherhand, the British pound was 9 percent undervaluedwith respect to the U.S. dollar, the Mexican peso36 percent, the Russian rouble 39 percent, the Chi-nese renminbi or yuan 42 percent, and the Indianrupee 61 percent. Norway was therefore the mostexpensive country for Americans to visit and Indiathe least expensive (among the countries listed inthe table).

(continued )

Salvatore c15.tex V2 - 10/18/2012 12:45 A.M. 8



■ TABLE 15.1. (continued)

Big Mac Prices

Actual Dollar Under (−)/Over (+)In Local In U.S. Implied PPPa Exchange Rate: against the

Currency Dollars of the Dollar Jan. 12, 2012 Dollar, %

Indonesia Rupiah 22.534 $2.46 5369 9160 −41Israel Shekel 15.9 $4.13 3.79 3.85 −2Japan Yen 320 $4.16 76.24 76.9 −1Malaysia Ringgit 7.35 $2.34 1.75 3.14 −44Mexico Peso 37 $2.70 8.82 13.68 −36Norway Kroner 41 $6.79 9.77 6.04 52Pakistan Rupee 260 $2.89 61.95 50.1 −31Peru Sol 10.0 $3.71 2.38 2.69 −12Philippines Peso 118 $2.68 28.11 44.0 −36Poland Zhoty 9.10 $2.58 2.17 3.52 −38Russia Rouble 81.0 $2.55 19.30 31.8 −39Saudi Arabia Riyal 10.0 $2.67 2.38 3.75 −36Singapore S$4.85 $3.75 1.16 1.29 −11South Africa Rand 19.55 $2.45 4.75 8.13 −42South Korea Won 3,700 $3.19 882 1158 −24Sweden SKr 41 $5.91 9.77 6.53 41Switzerland SFr 6.50 $6.81 1.55 0.56 52Taiwan NTS 75.0 $2.50 17.87 30.0 −40Thailand Baht 78 $2.46 18.58 31.8 −41Turkey Lire 6.60 $3.54 1.57 1.86 −16

a Purchasing-power parity: local price divided by price in the United States;b Average of four cities;c Dollars per pound;d Average of 5 cities;e Weighted average of prices in euro area;f Dollars per euro.Source: ‘‘2012 Big Mac Index,’’ The Economist, January 12, 2009.

Significant structural changes also lead to problems with the relative PPP theory. Forexample, the PPP theory indicated that the British pound was undervalued (i.e., the exchangerate of the pound was too high) immediately after World War I, when it was obvious that theopposite was the case (and the exchange rate of the pound should have been even higher).The reason was that the United Kingdom had liquidated many of its foreign investmentsduring the war, so that the equilibrium exchange rate predicted by the relative PPP theory(which did not take into consideration the drop in earnings from foreign investments) wouldhave left a large deficit in the U.K. balance of payments after the war. Case Study 15-3 pro-vides a simple test of the relative PPP theory. More formal and rigorous tests are discussedin the next subsection.



■ CASE STUDY 15-3 Relative Purchasing-Power Parity in the Real World

Figure 15.2 shows the relationship between changesin relative national price levels and changes inexchange rates for 18 industrial nations from 1973to 2011 (the period of flexible exchange rates). Thehorizontal axis measures the average inflation ratein each country minus the average inflation rate inthe United States (so that positive values refer toa higher average inflation rate in the nation thanin the United States). The vertical axis measureschanges in the foreign exchange rate, defined asthe foreign-currency price of the U.S. dollar. Thus,an increase in the foreign exchange rate refers to adepreciation of the foreign currency relative to theU.S. dollar, while a decrease in the exchange raterefers to an appreciation of the foreign currency.

Depreciation of foreign currencyrelative to U.S. dollar

3

2

1

–1

–1 1 2–2

–2

–3

–4

Appreciation of foreign currencyrelative to U.S. dollar

Italy

SpainNew Zealand

Australia

Foreigninflation rateminusU.S. inflation rate

U.KSweden

Ireland

Norway

Canada

Belgium

Netherlands

AustriaGermany

Japan

Switzerland

FinlandDenmark

France

FIGURE 15.2. Inflation Differentials and Exchange Rates, 1973–2011.Positive values along the horizontal axis refer to higher average inflation rates in the nation than in the United States.Positive values along the vertical axis refer to a depreciating currency relative to the U.S. dollar. Since nations with higherinflation rates generally experienced depreciating currencies the relative PPP theory seems to be broadly confirmedin the long run. Since 1999, changes in the exchange rates of EMU countries reflect the changes in the euro/dollarexchange rate.Source: International Monetary Fund, International Financial Statistics, various issues.

According to the relative purchasing-powerparity (PPP) theory, nations with higher inflationrates than in the United States should experiencedepreciating currencies, while nations with lowerinflation rates should have appreciating currencies.The figure shows that this is indeed the case overthe 38-year period examined. That is, countries withhigher inflation rates than the United States expe-rienced depreciating currencies with respect to theU.S. dollar, while countries with lower inflation ratesexperienced appreciating currencies. For the the-ory to hold perfectly, however, the plotted pointsin Figure 15.2 should fall on a straight line with apositive slope of 1. Since this is not the case, therelative PPP theory holds only approximately.



15.2C Empirical Tests of the Purchasing-Power Parity TheoryThe movement to a floating exchange rate system after 1973 stimulated a great resurgenceof interest in the purchasing-power parity theory and led to numerous empirical studies totest the validity of the theory.

Frenkel (1978) provided empirical evidence on the long-run validity of the PPP theoryduring the high-inflation years of the 1920s, and so did Kravis and Lipsey (1978) for the1950–1970 period, and McKinnon (1979) for the 1953–1977 period. On the other hand,Frenkel (1981) found that the PPP theory collapsed during the 1970s, especially in the latterpart of the 1970s, and so did Levich (1985) and Dornbusch (1987) for the 1980s.

Frankel (1986 and 1990) has suggested that researchers should utilize data overmany decades to properly test the PPP theory because deviations from purchasing-power parity die out only very slowly. Utilizing annual data on the dollar/pound exchangerate over the 1869–1984 period, Frankel showed that it took between four and five yearsfor one-half of the deviations from PPP to die out and that only about 15 percent of thedeviations from PPP were eliminated per year. Lothian and Taylor (1996), using datafrom 1790 to 1990 for the dollar/pound and the franc/pound exchange rates, confirmedFrankel’s results, as did Frankel and Rose (1995) using annual data for 150 countries from1948 to 1992, MacDonald (1999) using 1960 to 1996 data, Taylor (2002) using annualdata for 20 countries (the G-7 countries and 13 other countries) over the 1882–1996period, and by Cashin and McDermott (2002) for 20 industrial countries over the1973–2002 period. Taylor and Taylor (2004) review this empirical evidence and supportthe above results and conclusions. Cashin and McDermott (2006) extend and confirmtheir earlier conclusions for 90 developed and developing countries over the 1973–2002period.

Why do deviations from PPP die out so slowly? One possible explanation given byRogoff (1996 and 1999) is that, despite all the globalization that has occurred duringthe past two or three decades, international commodity markets are still much less inte-grated than national commodity markets. This is due to the existence of transportationcosts, actual or threatened trade protection, information costs, and very limited interna-tional labor mobility. As a consequence of various adjustment costs, exchange rates canmove a great deal without triggering any immediate and large response in relative domesticprices.

We can therefore come to the following overall conclusions with regard to the empiricalrelevance of the PPP theory. (1) We expect the PPP to work well (i.e., the law of one priceto hold) for highly traded individual commodities, such as wheat or steel of a particulargrade, but less well for all traded goods together, and not so well for all goods (whichinclude many nontraded commodities). (2) For any level of aggregation, the PPP theoryworks reasonably well over very long periods of time (many decades) but not so well overone or two decades, and not well at all in the short run. (3) PPP works well in cases ofpurely monetary disturbances and in very inflationary periods but not so well in periods ofmonetary stability, and not well at all in situations of major structural changes.

These conclusions are very important not only for the relevance of the PPP theory itself,but also because, as we will see in the rest of this chapter, the PPP theory occupies a centralposition in the monetary and in the asset market or portfolio balance approaches to thebalance-of-payments and exchange rate determination.

foreign exchange markets chapter and exchange rates markets/materiale... · 14.2 functions of the...

Documents